Home

R2R user manual - The Breaker Lab

1. 004 3 6 3 Automated multistem junction layout with direction constraints 3 6 3 1 Solving multistem junctions can be slow oe Internalloops a os ena AAA e A a EO QOO N O V o o o oo o Rh O OOO o RN 30 oe Fsendoknots cs crea dare 32 3 8 1 A note on the representation of pseudoknotted secondary structures within Stockholm fles s eai fo aot a SEEDER a a on ODER ORS E a 32 38 2 Pseudoknots drawn with callouts s lt a sas 54s dee ae ea eee ede eo 32 3 8 3 Pseudoknots drawn with in line style 34 3 9 Preparing presentations using projectors e a e a a 36 3 10 How to draw GOLLD RNA from R2R output aaa a a 36 oll Making the demo output iles so rec p ema WE ee PEE ewe EE a a 36 Reference automated inference of nucleotide conservation levels 38 4 1 Recommended command 2 44554 ers erase ses 38 42 Fragme ntary alignments lt o oaa o sasaa rerne ee RR erraren 39 4 3 Weighting sequences using your own algorithms 39 Ad Generalcommand s so s aora doci E ahe a a n O A A N 40 4 5 Output of R2R for the consensus 464644485 drasa Grr rrasa eS 40 4 6 Generating your own alignment consensus bypassing R2R 41 Reference drawing 42 al Rommi ROR sio pa e e we e dos aa ae is a 42 52 About the Stockholm ile torial s cs 4 624084 6 4 049 bee eR ER OSS RH 42 5 2 1 A note on R2R s representation of consensus second
2. lt lt a 69 5 8 6 9 Shading nucleotides along the backbone 69 5 8 6 10 Outlining a stretch of nucleotides around both ends 69 5 8 6 11 Shading the backbone in skeleton drawings 70 5 8 6 12 Drawing circles associated with loops 70 5 8 6 13 Drawing direct lines between consecutive nucleotides 70 SST Miscellaneous s para ac A ee AE AA A eG 70 5 8 7 1 Override default parameters for drawing 70 Oe No5 prime label score A A 70 58 7 3 Adding G lor transcription cs ae es ewe ee RR EO 71 5 8 7 4 Keeping gap columns in the drawing 71 Boot Breaking US ic ke riada Later AAA 71 5 8 7 6 Deleting columns using an explicit command 71 Troubleshooting AI 71 Text output of r2r useful in debugging cov rior Ga REE SS 72 310 1 Starend ol Mle AMA eidon trta Genies 72 5 10 2 consensus lines and raw coordinates 72 5 10 3 Preliminary secondary structure drawing units 72 A J04d place emi inke o soss oe ERROR a EER EES 73 5 10 5 Selecting the place explicit commands 74 nll Interacting with ERE using SCripls ss o oec os ed ee eee aa a h 74 5 11 1 Generating R2R input using seripte lt lt lt lt 4 405 74 5 11 2 Accessing layout information from R2R using scripts 75 6 Reference modular structures TT 2 1 AA A Re ee 77 6 2 Using
3. Rcau The same RNA structure depicted in Figure 3 16 is now drawn with an in line style pseudoknot A R2R s default drawing of the structure R2R does not attempt to make a feasible in line style drawing of a pseudoknot structure since this is quite difficult and sometimes not possible within R2R s quality constraints B Stockholm input file This is the contents of demo demo pknot inline sto Some commands were removed but these lines are not necessary to generate the drawing in part C C Raw output of R2R when run on the Stockholm file in part B D Output of R2R when run using the command R2R SetDrawingCommand showPlaceExplicit true which marks where place_explicit commands were used The boxed region is shown at a larger scale in the next subfigure E Expansion of the region boxed in part D The pink arrow labeled 11 on the top left corner refers to a place_ explicit command on line 11 of the input sto file demo demo pknot inline sto The pink line has two arrowheads The thicker one pointing up and right indicates the direction in which layout actually proceeded although it is not usually necessary to know this The gray arrow on the left labeled def shows a default rule that R2R used to position these nucleotides In this case the default rule says that a helix should be positioned in a linear arrangement These arrows are sometimes useful to diagnose problems with place_explicit commands The arrows and thei
4. pairBondCircleLineWidth width of the line of the circle used for G U or non canonical base pairs Default 0 002in minPairShadeGap minimum distance between shading rectangles of consecutive base pairs When base pairs are drawn at an angle 45 degrees is worst the shading box s width needs to grow in order to cover the whole nucleotide letter since nucleotides are rectangles that are always drawn in the same orientation Eventually these shading rect angles can get very close or even overlap which looks bad Therefore R2R will cheat 46 and allow some small parts of nucleotide letters to be unshaded in order to guaran tee a minimum distance between the shading rectangles and this minimum distance is minPairShadeGap Default 2 5pt e Circles that represent nucleotides whose identities are not conserved anyNucCircleWidth the width of the circle s line Default 0 O1in e Variable length stems or loops varHairpinNumFakePairs for the var hairpin command the number of fake base pairs to draw Integer Default 3 varTerminalLoopRadius for the var hairpin or var_term_loop commands the radius of the arc used to represent the terminal loop Default 0 17in e Shading along backbone or drawing in skeleton mode outlineAlongBackboneWidth width of shading for outline along backbone com mand default 0 7pt shadeAlongBackboneWidth width of shading for shade_along backbone command
5. AdobeGraphicsLayout h Header file for above AdobeGraphicsPdfLike cpp Generic class to factor out similarities between drawing in PDF and SVG AdobeGraphicsPdfLike h Header file for above Cm2HmmOptimize_cfsqp cpp Implements SolverWrapper interface see Optimize h for CFSQP File is originally from my earlier rigorous filter work CommaSepFileReader cpp Reading of files delimited by a single character e g commas It is used to read tab delimited r2r meta files and space separated sto files CommaSepFileReader h Header file for above GSCConsensus cpp Implements the GSC weighted consensus functionality The GSC algorithm itself is implemented by the code in NotByZasha infernal 0 7 squid LabelsAndProjection cpp Functions related to parsing sto files related to projection of columns in an alignment or labeling of columns MiscExceptions cpp Generic exception classes for error handling MiscExceptions h Header file for above NoUnderflowDouble h Implements real numbers that have a greatly expanded range for their exponent Used by code related to CFSQP Optimize cpp Generic functions for non linear optimization Also implements most of the functionality needed for the solver cache 84 Optimize h Header file for above ParseOneStockholm cpp Functions for interpreting R2R commands in a sto file Note parsing happens in three passes Pass 1 in which the alignment and column annotation da
6. Y R GO stem 0 e lt multistem_junction_circular M allstems any angle draw_circ U R R y O li OR tao 3 nt M Y R Ca 597C e c G D Une ESE cre E c Lue s C Y a Cc cGy aC Gre O A Cdr Pry Ere Ce Y 0 YaR ACI 6 0 O G R 0 G G YA U G Y Cc S G Ac O GY A Default layout of a hypothetical three stem junction that is modeled on a multistem junction in HEARO RNA 15 This output is based on the files demo demo multistem sto and demo demo multistem r2r_ meta B In the default layout nucleotides along the multistem junction are positioned on a circle The R2R command shown here directs R2R to draw this circle The command refers to position M Position M is labeled and is the left nucleotide of the base pair that encloses the whole multistem junction Note the command should be one line but is displayed on multiple lines in this figure so it fits the page C Stems can be flipped so that they are directed into the multistem junction This can save space when the multistem loop is larger than one of the stems The stems are numbered starting at zero The enclosing stem does not have a number and cannot be flipped D The SAM IV riboswitch 16 is drawn using the default layout of its multistem junction Note the gray lines correspond to pseudoknots Drawing of pseudoknots are explained in a later section of this tutorial 23 Figure 3 9 Manual layout of multistem junctions
7. e Use the tick_label command to label nucleotides and make sure you have the right one or find out where it is like GF R2R tick label regular numbering 1 1 zero e cyclic dependencies very rare try using ignore ss except for pairs SS_cons code outline for any extra SS_cons lines and see if the problem recurs For example if you have GC SS_cons_1 use ignore_ss_except_for_pairs _1 outline 71 e mark_flip This R2R command marks nucleotides that are flipped left vs right with f and un flipped nucleotides with Only the 5 most nucleotide in each structural unit stem bulge etc are marked e You can get R2R to show extra information about how it s using place_explicit commands See section 3 8 3 for examples and use the following command GF R2R SetDrawingParam showPlaceExplicit true e You can get R2R to show you the direction of the backbone at each nucleotide using GF R2R SetDrawingParam showEachNucleotideDir true 5 10 Text output of r2r useful in debugging The r2r program produces output that might help in debugging problems with your input or indeed the R2R code Sorry there s no quiet option 5 10 1 Start end of file For each Stockholm input file R2R will print PROCESSING file name When it finishes parsing and layout of the RNA it will print DONE file name Note that the actual drawing code is performed after parsing layout of all structures and some error conditio
8. e multistem_junction_circular_solver e multistem_junction bulgecircley_solver e multistem junction bulgecircleynormals solver All _solver variants use a very similar syntax and framework They are explained in full detail in Section 5 8 3 2 but for now we will see the most essential aspects of the commands Examples of the solver commands are given in Figures 3 10 3 11 3 12 and 3 13 The command begins with a solver command then specifies a label As with previous multistem junction commands the label denotes the left nucleotide in the base pair that encloses the multistem junction Then then properties of each stem in the multistem junction are written The enclosing stem is named sO then each stem from 5 to 3 are labeled s1 s2 see Figure 3 10 The names for the stems are different from those of multistem junction commands described earlier in this tutorial In particular the enclosing stem can be explicitly referenced as s0 For each stem the desired angle of that stem is given in a coordinate system defined by the enclosing stem Figure 3 10B The point at which the circle should intersect the inner base pair of each stem is also specified Useful values are m intersects the midpoint of the base pair 1 intersects the center of the left nucleotide in the base pair r intersects the right nucleotide in the base pair and ai automatically solve the intersection To find a good layout of a particular multistem junct
9. this requires that you have CFSQP Note that this process will take a considerable amount of time possibly hours because several RNAs use the solver and the solver sometimes requires significant amounts of time to obtain optimal solutions Tip if you have GNU make and have a multiprocessor machine you can use the j flag For example if you have 4 CPUs and want to make PDF files run make j 4 all pdf You can determine if you have GNU make by running make version GNU make will say GNU Make on the first line 37 Chapter 4 Reference automated inference of nucleotide conservation levels In order to draw a consensus diagram it is necessary to determine which nucleotides are conserved i e are part of the consensus sequence It is also useful to determine how conserved they are and to annotate which base pairs exhibit covariation R2R can perform this calculation and this output is subsequently used by R2R to actually draw this consensus information This process of inferring a consensus then drawing it was illustrated in Chapter 3 To draw a single RNA molecule it is not necessary to determine a consensus The definition of these consensus statistics was defined in a previous report 13 which read To establish the extent of conservation reflected in consensus diagrams sequences were weighted to de emphasize highly similar homologs Weighting used the GSC algorithm 3 as implemented by Infernal 2 and weight
10. e GC conss The degree of conservation 1 is the highest level of conservation When the consensus letter in GC cons is a nucleotide or degenerate nucleotide by default the levels range from 1 to 3 and represent how often that nucleotide is found in this alignment position When the consensus letter is lower case n by default the levels range from 1 to 4 and represent how often any nucleotide is found in this position as opposed to a gap When the consensus letter is a dash representing a gap the number is always O zero An annotated example is given in Figure 3 1D e GC cov_SS_cons Covariation annotation 2 represents base pairs that show at least one instance of covariation 1 means there is at least one compatible mutation 0 zero means there are no mutations so no data means that there are too many non canonical base pairs An annotated example is given in Figure 3 1D e GC cov_SS_consz covariation data for pseudoknot structure line GC SS_consz e GF NUM_COV Number of base pairs with covariation per file annotation e GF WEIGHT_MAP Lists GSC algorithm weights used for each sequence in a space spearated list The even numbered fields identify a sequence and the odd numbered fields are the weights per file annotation Note if you use GF USE_THIS_WEIGHT_MAP the results of GF WEIGHT_MAP will be identical to the weights you supplied e GF USED_GSC TRUE Confirms that R2R used the GSC algo
11. Stockholm format The key ideas is that R2R uses the following types of annotation e GF tag data This specifies file level markup For example R2R drawing commands use the form tag R2R GF R2R command 42 e GC tag columns In this case each position in the string columns corresponds to each column in the alignment For example for tag SS_cons the symbols in the columns string specify the consensus secondary structure of the alignment e GR hitld tag columns Each position in the string columns corresponds to each column in the sequence specified by hitId see Section 5 4 1 This is relevant to identifying columns in an individual RNA when the user wants to draw that RNA sequence instead of the consensus Pedantic point R2R s input format does not quite respect the Stockholm format For example it uses many GF R2R lines each of which represents a distinct command However the Stockholm format says that these lines are logically part of the same line and the separation of commands is technically not enforced However has not been a problem in practice 5 2 1 A note on R2R s representation of consensus secondary struc tures R2R uses a non standard method for writing consensus secondary structures for motifs with pseu doknots To describe pseudoknots in addition to the standard GC SS_cons line there can be other lines of the form GC SS_consX for any string X typically I call them GC SS_cons_1
12. applied with one of the following three related commands 24 Figure 3 10 Automated layout of multistem junctions A U R B zero degrees C mme K negative angles JN 7 positive angles E o ME O R 3 nt O e 0 3 nt 0 0 s1 11 Y R Y R o G C SC 3 nt 5 5 multistem_junction_circular_solver M s ai 0 s1 90 ai 0 s2 0 ai 0 draw_circ multistem_junction_circular_solver M sO m s1 90 m s2 m draw_circ O 2 O Q Q a loa E cnc C G e Gy CESS ue cme C ds ame C G CG 0 C G A JG Aec G A A AS U U Y Ce U Ac G a Ac G A RCo mua G C cay Ga C e ODIO lt lt I lt O000 DoE EES I lt 0000 keee O N Ccu ccucryc ie ES OMe uii Oe ee Ge a a Co GA CGACYR neo b Ko a G U x lt A O y lt O O 7 Seo gt O EG E n a U A G C ece G C CGG G C Latins E og 5 GGYY multistem_junction_circular_solver 1 multistem_junction_circular_solver 1 s ai 0 s1 90 ai 0 s2 O ai 0 s 0 ai 0 s1 45 ai 0 s2 ai 0 draw_circ draw_circ A Layout of the demo structure used in previous figures using the solver The solver command is given below the drawing The lower case m parameters say that the circle should intersect in the midpoint m of stems Note that the length of the variable length region marked as 0 3 nt is adjusted by the solver to optimize the circular layout The drawing is based on the file demo demo multistem sto with solv
13. straight line gap between them will be a bit smaller than alongBackboneMidpointGapLength Second R2R draws the backbone shading as a line arc and sets the thickness of the line to simulate the shading The length of the gap is thus calculated at the midpoint of the shading line But if the path is on a circle the two ends of the shading line will not be of equal distances Usually the circles are of a large enough radius that these effects have no practical effect By the way I implemented this feature to experiment with shading nucleotides involved in a pseudoknot base pair to show covariation as an alternate way of indicating pseudo knots These pseudoknotted nucleotides are often on a circular path The gaps between shading are meant to simulate the gaps between consecutive nucleotides in a regular 47 stem Some other changes would be needed to implement this style especially since cur rently all shade_along_backbone commands use the same drawing style so covariation shading and other user directed shading would conflict If alongBackboneStyle is not set to 1 this parameter has no effect Default 0 i e don t leave any gap e Debugging help showPlaceExplicit Boolean true or false Shows information on what place explicit commands were used to position drawing units shown as thick lines as well as unused commands thin lines The lines connect the two nucleotides that were positioned based on a place_explicit
14. 6 SvgGraphics cpp Drawing into SVG files Concrete implementation of AdobeGraphics interface SvgGraphics h Header file for above SymbolicMath cpp Implements symbolic expressions used for CFSQP based solver of mul tistem junctions SymbolicMath h Header file for above UseDebugNew h Header file used in debugging multiDimVector h STL style classes for multi dimensional arrays stdafx h Include standard system headers vectorPlus h Extension of STL vector class to detect array bounds errors 85 8 2 Other information 8 2 1 Overall layout of RNA To position the nucleotides in an RNA structure the structure is first decomposed into struc tural elements consecutive base pairs internal loops or consecutive unpaired nucleotides Default constraints define each element s orientation relative to another element To apply user defined con straints in an intuitive manner a graph is defined whose nodes are structural elements and whose edges correspond to constraints This graph s minimum spanning tree with user defined constraints prioritized over default constraints yields the positioning constraints to apply However some rules direct nucleotides to be drawn on a circle connecting two already positioned nucleotides and these rules are applied after the nucleotides are positioned Constraints in the graph correspond to place_explicit commands This drawing order logic is applied after any multistem_ comma
15. A a o B multistem_junction_bulgey 1 se J 45 18 3 2 90 C Gy J1 451000 90 U c c C G 7 CES AC G E G C U Ap 9 emme l GG G CC ue ema O eo Ae Q C7 PCUPGCUGRYC G R 0 U Cc ee CI U G Y Y RG JO G cc GA CGACYRG G E YA AR YA A G A Ye G e Y G O YAC G G YAC G U A ah G C CGG U A 111111 5S Vd G C A The published layout of SAM IV riboswitches 16 Layout of the multistem junction uses the multistem_junction_bulgey command This drawing is based on the file demo SAM IV SAM IV sto in multistem original mode This mode is entered in the r2r_meta file using the specification define multistem original Note at the time at which SAM IV riboswitches were drawn manual layout was the only option for drawing multistem junctions Also the gray lines are for pseudoknots as described in Section 3 8 2 B R2R command to draw the multistem junction and larger view of the multistem junction The junctions called JO and J1 are labeled There are no nucleotides within junction J1 In the multistem junction bulgey command the stems within the multistem junction are positioned using functionality similar to place_explicit commands The junctions are then po sitioned automatically between the stems by positioning them on circles equivalent to the bulge command Figure 3 9 shows an example Junctions are identified using the symbols JO J1 J2 etc Junction JO identifies the junction between the enclosing stem and
16. Added ability for user to weight sequences using an arbitrary method instead of the GSC algorithm while calculating frequencies otherwise as before See Section 4 3 and Section 7 2 2 6 e Version 1 0 1 distributed with revised version of paper November 2010 Added demo demo c di GMP 11 c di GMP II update sto re drawing based on new sequences for Wikipedia article also some files in the demo pedagogical directory Added Additional files 3 and 4 from the paper to software distribution Added feature indicateOneseqWobblesAndNonCanonicals added as a drawing param eter Fixed bugs x Fixed a problem when internal loops are converted to bulges explicitly using the internal_loop_to_bulges command or implicitly by applying a bulge or place_ explicit command to a nucleotide within the internal loop and when the user did not specify what to do with both sides of the internal loop x Fixed various issues with applying the multistem_junction_circular command to 3 bulges Formatting issues with the user manual The GSC weighted consensus flag was explained in more detail e Version 1 0 distributed with initial submission of paper July 2010 Chapter 2 Installation Installation essentially means making the executable file r2r 2 1 Platforms on which R2R is known to work Building of R2R has only been tested with the gcc compiler suite http gcc gnu org To ensure compatibility you will need the GNU
17. Boolean or operation which evaluates to the true if either of the sub expressions are true The exclamation mark is the Boolean not operation so the expression predValue GNRA represents all sequences that do not match the GNRA predicate Warning if you use Boolean predicates you must define predicates they use within predValue earlier in the sto file since R2R will evaluate predicates in the order in which they appear in the sto file The script SelectSubFamilyFromStockholm pl evaluates predicates and creates a new sto file with the subset of sequences It also modifies which commands are present in the file in order to use commands that are appropriate to drawing the modular structure All lines beginning with GF R2R or GC R2R are by default removed At the same time for example with the predicate GNRA lines beginning GF SUBFAM_GNRA_R2R are changed to GF R2R so that they now become processed Similarly lines beginning GC SUBFAM_GNRA_R2R are changed to GC R2R Chapter 6 gives further details on the commands 3 8 Pseudoknots Pseudoknotted RNAs typically lead to confusing drawings when drawn with the default layout of R2R Therefore to help you to plan the final layout better the first thing to do with an RNA structure that has a pseudoknot is usually to make R2R ignore the pseudoknot which can be accomplished with the ignore_ss_except_for_pairs command I give an example of this command in the next section
18. C compiler which provides the gec command and the GNU C compiler which provides the g command R2R has been tested on the following platforms e gcc version 3 4 4 under Cygwin 1 5 25 32 bit on Windows XP e gcc version 4 3 4 under Cygwin 1 7 1 32 bit on Windows 7 e gcc version 4 1 2 under Ret Hat Linux running Linux kernel 2 6 18 64 bit e gcc version 3 3 under MacOS s Darwin version 8 8 0 32 bit I presume that R2R will work on gcc version 3 or higher on Cygwin Linux or MacOS Darwin R2R is known to produce output that can be used with Adobe Illustrator CS on Windows XP Inkscape 0 46 on Windows XP and CorelDRAW Graphics Suite X4 on Windows Vista I presume however that the output will work on any version of these programs at least with appropriate setting of fonts see below 2 2 Decide whether to install CFSQP or not R2R implements multiple methods to find a good layout for multistem junctions Some automated solutions require software that can solve non linear optimization problems You have two options e Option 1 Do not install a solver This is the default Advantage easiest installation Disadvantage some functionality for multistem junctions won t work If you do not use the multistem junction non linear solvers there is no disadvantage to this option However the R2R commands multistem_junction_circular_solver multistem_ junction_bulgecircley_solver and multistem_junction_bulgecircleynormals_sol
19. LIEIL e LL 1011 A CGACYRG RCGACYRG YeR Y This drawing is based on SAM IV riboswitches 16 13 in the file demo 22 SAM IV sto A An R2R command to turn an internal loop in SAM IV riboswitches The label K refers to the 5 most nucleotide within the 5 part of the internal loop as shown You can also use the 3 most part of the 3 part of the internal loop The stem can be turned 90 degrees clockwise or counter clockwise The value 1 as shown is clockwise while 1 would be counter clockwise R2R will decide the layout on one side of the internal loop The value r will make R2R use the right side i e 3 side for positioning while 1 is for the left side B The turned internal loop in the context of the full SAM IV consensus C E Turning the internal loop using solver commands Since the solver commands require the label to be the left nucleotide of the enclosing base pair of the internal loop we use the label K Therefore I have conservatively set the solver to use up to 100 000 iterations You can decrease this number with an R2R command like SetDrawingParam solverMaxIters 50000 I have also implemented a caching strategy in which solutions to problems solved by CFSQP are store in a file that is associated with the r2r_meta input file These files end in the extention solver cache If you re run R2R but do not change anything related to the multistem junction the solution will be re used By deleting the solver
20. Note that the multistem_junction_bulgecircley commands do not explicitly deal with circle intersection and this parameter e g ai is ignored However it must still be supplied After the label is specified additional commands are in a list and each command element is as follows e si angle strategy You must provide one of these options for each stem The computer will tell you if you don t Tells the computer how to position stem 7 The stems are defined differently for the two implementations For solver stem 0 is the enclosing stem of the multistem junction stem 1 is the first stem clockwise after the enclosing stem and so on going around clockwise or equivalently 5 to 3 For multistem_junction_circular the enclosing stem has no number and always fits perfectly on the circle and stem 0 is the first stem clockwise from the enclosing stem angle is the direction in which the stem points relative to the left nucleotide of the enclosing stem For example angle 90 means that the stem will be pointing clockwise by 90 degrees relative to the angle of the enclosing stem In the usual layout the enclosing stem is pointing up the page from the outside of the stem pointing into the multistem junction Thus angle 90 is pointing left angle 0 is also pointing up the page angle 90 is pointing to the right In the solver commands setting the direction of stem 0 the enclosing stem has the effect of rotating the other s
21. SelectSubFamilyFromStockholm pl o 77 il Command ines r ss koa we EER hee PR ee POR ee Pe ea ees TE 6 2 2 How to define predicates lt lt e os TT 6 2 2 1 Regex predicates lt lt o re es 78 Maca REL Ose rocio E Aaa a a 78 6 2 3 How the file is modified by applying a predicate 79 7 Meta Makefile 80 fol UDS the SUI os canta da AAA ARA HOS 80 TLI eC options ARI 80 Tala Makelle arpels sica ada ES ERE ES RR RS 81 Ta Additional markup for ato fee s 6b dw Oh aE SY AG eR OR ER DESO 82 Tal Ampere rales 22 kG bet SERRE REA ORS wR OEE a e 82 Vee ON ee a te ee eee eww eee SE E 82 AAA eh ERG RES EG REE e SHG ES 82 Taba Es 00 eee ae ee AAA eh eee we ES 82 7 2 2 3 Defines for consensus diagrams o 82 7 2 2 4 Using define directive with modular structure 82 7 2 2 5 Oneseq for a predicate defining a modular structure 83 7 2 2 6 Weighting sequences using your own algorithms with MetamakeDemos 83 8 R2R source code 84 8 1 Summary of C and Perl source code files 0 0 00 0008 84 o oe ss So oe ee ey oh ae eo ee ew ee ee ee 86 dl Averell layout o RNA socorrista aer TORE MA 86 822 Interring a path for the backbone o se 2608448 ee ae bonbe noo 86 Chapter 1 Introduction 1 1 What does this software do For a full description of this software please read the paper titled R2R softwar
22. User specified place_explicit commands are shown in magenta while default rules are gray Lines are labeled Default rules say def The positioning of the 5 most nucleotide says lst User specified place_ explicit commands are labeled with their line number in the input Stockholm file To simply the layout I used a very small font size to label the lines You will probably need to use display software to zoom into the picture in order to read the labels Default false showEachNucleotideDir Boolean true or false Shows the direction vector at each nucleotide Useful in particular for working out place_explicit commands Note that some nucleotides have two directions These are nucleotides that are positioned in a straight line on their 5 direction but are on a circular layout on their 3 end Linear directions are shown with a normal arrow Circular directions are shown with a line that has a small circle on the direction end Direction vector lines always originate at the center of the nucleotide Lines are drawn in cyan Default false dumpInfoFile if non blank the name of a file into which R2R should dump information on the layout and other features See section 5 11 2 If blank no information is output Default blank e Parameters relevant to skeleton mode skeleton_scaleMeasurementsBy alternate value for scaleMeasuresBy to be used in skeleton mode Default 0 25 skeleton_pairBondWidt
23. a subsequence The asterisk represents zero or more occurrences of the previous symbol Thus the expression matches all subsequences that consists only of gap characters 30 Figure 3 15 Modular structures A B STOCKHOLM 1 0 one GAGAAA UCAACCCUUGGGGGAGCA two asuuc cuaauccuaccaccacca C amp MO modular demo modul GENRA ee 00 three GGGAAACCCAA GGAGCA LY LO ae four AGCAAC CUAA GGAGCA G C 0 0 AA GC SS_cons Vine A eee A O Guu GC R2R_LABEL occccnonananananananrooo GC SUBFAM_LABEL_TERM o be Pad eae A E z A GC SUBFAM_LABEL_OPT s cs se eee eee red demo medular oer demon me qulaNCG GC SUBFAM_GNRA_R2R_LABEL 50 27 GC SUBFAM_UNCG_R2R_LABEL Je ee GC SUBFAM_OPT_R2R LABEL C G G C Y R GF SUBFAM_REGEX_PRED HAS_GNRA TERM G A Z AG A H GF SUBFAM_REGEX_PRED HAS_UNCG TERM U A Z CG GF SUBFAM_REGEX_PRED FOUR TERM A Z A Z A Z A Z H GF SUBFAM_PERL_PRED GNRA return predValue HAS_GNRA 88 predValue FOUR GF SUBFAM_PERL_PRED UNCG return predValue HAS_UNCG 88 predValue FOUR H GF SUBFAM_REGEX_PRED OPT OPT A Z GF SUBFAM_GNRA_R2R no5 C re ae GF SUBFAM_GNRA_R2R set_dir pose 90 eerie allu G GF SUBFAM_UNCG_R2R no5 c e A U 6 e GF SUBFAM_UNCG_R2R set_dir pose 90 TU AMAIGGAGCA GF SUBFAM_OPT_R2R no5 50 GF SUBFAM_OPT_R2R set_dir pose 90 ye C
24. and then having the computer render the single stranded junctions as bulges It is largely equivalent to a series of place explicit and bulge commands The main benefit is that you only have to create one label i e label Note that multistem_junction_bulgey also provides some seldom used functions that do not use bulge lay outs for single stranded regions label specifies the left nucleotide of the base pair that encloses the multi stem junction For examples see Section 3 6 R2R will complain if you specify something that is not the left nucleotide of an enclosing pair By default each junction within the multistem junction is laid out as if a bulge command were used However you can specify alternate layouts as explained below After label is a series of commands Each command is one of the following 57 e Ji anglePlace x1 y1 x2 y2 angleMove where J is the literal letter and 7 is a number specifying which stem The enclosing stem is assumed to have a known position Then the next stem is positioned relative to the left nucleotide of the enclosing stem pair This next stem is positioned using JO Then the next stem is positioned using J1 relative to this stem This continues until we ve done all stems immediately enclosed by the enclosing stem Basically we re going around the multi stem junction from 5 to 3 The rest of the Ji command is the same as the place_explicit command For JO we position the left nucleotide of th
25. are two command lines you can use to process Stockholm format alignments 1 r2r input file sto output file where output file will be created in either Adobe Acrobat PDF format or Scalable Vector Graphics SVG format see below and input file sto is a file in Stockholm format Note some functionality is not available in this command line format so option 2 below is preferable 2 r2r input file r2r_ meta output file where output file will be created in PDF or SVG format and input file r2r_meta is a file that lists individual Stockholm format files and has additional options Its format is described in Section 5 5 The diagrams drawn will be organized into a grid in output file The output file name must end in either pdf or svg which determines what output format R2R will write Note PDF output is optimized for Adobe Illustrator while SVG output is optimized for Inkscape mainly in terms of the font Of course with SVG output you can easily do a search amp replace to change the font However R2R internally knows approximately font metrics which allows it to position text almost accurately Therefore you should use a font that is related to Helvetica Helvetica Arial Bitstream Vera Sans or DejaVu Sans 5 2 About the Stockholm file format R2R uses the generalizable format of Stockholm files to specify markup that defines how a mo tif is drawn Stockholm format is explained on Wikipedia http en wikipedia org wiki
26. cache file you can force a re computation Note R2R is designed to detect subtle changes that can change the problem and R2R will re compute the solution in these cases For example if you change the value of internucleotideLen using the SetDrawingParam command this changes a parameter in the multistem junction Non Linear Program even though it is an indirect effect 3 6 4 Internal loops Internal loops and bulges are unpaired regions that interrupt stems It is often desirable to turn the direction of the stem at these loops for two main reasons First bulges and highly asymmetric internal loops can have a distorted look due to having too many nucleotides on one side A turn can give more space to the longer side improving the clarity and aesthetic quality of the drawing 29 Second turning of stems can resolve problems with other regions of the structure that might overlap use space inefficiently or otherwise be undesirable R2R has a special command for turning the direction of the stem at internal loops or bulges This command restricts turning to 90 or 90 degrees however The command turn_stem_at_ internal is demonstrated in Figure 3 14 and documented in detail in Section 5 8 2 5 The turn_stem_at_internal command is limited both in the angles it can handle and in the layout policy You can use the _solver commands that are designed for multistem junctions on internal loops or bulges An internal loop is simply a multistem
27. code comprising R2R since the R2R program uses zero based numbers internally 67 5 8 6 2 nobpannot nobpannot Do not shade any base pairs Normally R2R will shade base pairs in consensus diagrams to annotate them as covarying compatible etc 5 8 6 3 Pseudo bold fonts thick_stroke font label stroke width Draws nucleotides matching label with stroking which has the effect of making them look bold R2R cannot select bold fonts directly Note this feature has not been implemented with SVG output It only works with PDF 5 8 6 4 Changing nucleotide colors nuc_color label color Sets the font color of the nucleotide s specified by label 5 8 6 5 Circling nucleotides circle_nuc label color width width in points Draws circles around all nucleotides with label The style of the circling is like for cleavage diagrams The optional width directive specifies the thickness of the line forming the circle By default it s the same as for the cleavage diagram notel To accommodate the circles the size of nucleotides is reduced by a constant factor defined by nucShrinkWithCircleNuc default 0 8 and the size of bond lines by a factor of pairBondScaleWithCircleNuc default 1 so no actual change To eliminate this behavior or find alternate measurements that work use the SetDrawingParam command see Section 5 5 2 5 8 6 6 Boxing nucleotides box_nuc label color width width in points Draws boxes rectangles aroun
28. direction is not orthogonal to the X or Y axis C Alignment constraints are added to the layout of part B align_angle 30 0 4 dictates that the stems sO and s4 should be vertically aligned The alignment angle 30 is used because the enclosing stem is rotated by 30 degrees see part D The constraint align nuc_centers_angle 60 align X align Y forces the midpoint of the nucleotides defined by align X to be horizontally aligned with the midpoint of the align Y nucleotides This alignment further improves the symmetry Note that R2R cannot aestheti cally satisfy any arbitrary constraint and so stems s1 and s3 cannot be vertically aligned in this layout D The drawing of part C is annotated The nucleotides that were aligned are drawn in alternate colors as shown The align X nucleotides were vertically aligned with each other when their stems sO and s4 were aligned Zero degrees is defined as the direction of stem sO as shown so the vertical direction is 30 degrees The solver directives draw_circ and draw_zero_degrees were added to draw the multistem junction circle and the zero degrees mark and nuc_color was used to color the align X and align Y nucleotides 26 Figure 3 12 Automated layout of multistem junctions additional example A GAA B GAA C A Cc A Y R Y R Y R Y R g cCMERcC o oant OA 0 2 A E 7 E o A ae A A G A A AO A A e RReeee a puami F e theese o Q 110000 e ArGA A y ARGA A Aa e A ABONO CA au CAcec
29. et al 2008 11 e demo 104 Weinberg et al 2010 17 e demo exceptional Weinberg et al 2009 15 Note The RNAs depicted in Supplementary Figure 11 of this paper are primarily described in the Weinberg et al 2010 paper 17 and are therefore found within the demo 104 directory The drawing of GOLLD RNA is somewhat complicated by different degrees of conserva tion of the structural elements For instructions on reproducing the drawing of GOLLD RNA in Fig 2a see Section 3 10 Also a careful look at Fig 2a in Nature will reveal that many nucleotides are slightly misaligned The reason is that the R2R drawings in this case were done using Adobe s Myriad font which is narrower than Helvetica The journal changed the font but did not realign the nucleotide letters The output of R2R is always aligned correctly The HEARO RNA skeleton drawing in Figure 3b was not drawn with R2R However an R2R drawing of this style is included Figure 3b was drawn before I had implemented skeleton drawings in R2R e demo c di GMP II Lee et al 2010 6 e demo Moco Regulski et al 2008 9 New drawing based on original layout e demo SAH Wang et al 2008 12 New drawing based on original layout e demo SAM IV Weinberg et al 2008 16 e demo hammerhead Perreault et al 2010 7 e demo additional Drawings of RNAs that were identified by other groups The tRNA drawing is based on the standard la
30. for var backbone_range units in the junctions The length units are angular and 1 unit is equal to the rotation in angle caused by the default internucleotide length with the given circle radius The output will also show any constraints and the value achieved for them 5 8 4 Changing layout of secondary structure 5 8 4 1 depair depair label label Assumes that label is the left nucleotide of a base pair Removes the pairing Useful for when one side of the pair should be a gap but not the other side label can also specify a set of left nucleotides like with SS_cons lt Multiple labels can be given separated by spaces For example to remove all the pairs in a pseudoknot corresponding to line SS_cons_1 use depair SS_cons_1 lt 5 8 4 2 make_pair make_pair left label right label ss name Makes the nucleotide left label base pair with the nucleotide specified by right label by altering the secondary structure By default the alteration will happen in the primary secondary structure i e SS_cons but the optional ss name can specify an alternate secondary structure line to use In theory you can specify multiple nucleotides in each of left label and right label and the corresponding pairs in reverse order for right label will become base pairs Thus again in theory you can specify a stem at once However in practice I haven t tested this functionality 64 5 8 4 3 Internal_loop_to_bulges internal_loop to
31. in a given paired position The command will create a file output sto that will have the same content as input sto but will contain additional markup summarizing what is conserved In the output sto file the line GC cons will contain a dash for columns that are classified as gaps and a nucleotide for non gap columns Gap columns are by default removed from the consensus when R2R draws it See Section 4 5 for details on the added contents of the output sto file 4 2 Fragmentary alignments The fragmentary feature described in this section was motivated by GOLLD RNA 15 GOLLD RNAs average roughly 800 nucleotides in size but are predominantly found in metagenome data from Lake Gat n that consists of short sequence scaffolds Therefore most of the detected GOLLD RNAs are fragments of the full RNA structure Moreover since relatively few GOLLD RNAs are currently available it is not practical to discard fragmentary GOLLD RNAs as this would greatly reduce our ability to analyze its structure To handle this situation R2R has a fragmentary mode In this mode for example any sequence that consists entirely of gap characters on its 5 end is assumed to be a fragment that is cut off at this 5 end In this case nucleotides that are expected to be 5 to this cut off point are treated as unavailable data As such they do not participate in any of the conservation statistics calculations neither in the degree of con
32. junction with only 2 stems for the purposes of these commands The application of solver commands to internal loops is demon strated in Figure 3 14 Note that R2R does not distinguish between internal loops and bulges as RNA biologists often do For R2R a bulge is simply an internal loop with zero nucleotides on one side 3 7 Modular structures I define modular structures as sub structures of an RNA that are sometimes but not always present in a conserved RNA structure Many RNAs for example have stems that are sometimes present To draw a consensus it is often desirable to reflect these modular structures and their relative frequencies As with other frequencies R2R weights sequences using the GSC algorithm and calculates the weighted frequency of the set of sequences that do have the modular structure See Chapter 4 for details Figure 3 15 demonstrates the drawing of two modular substructures in the context of a simple hypothetical motif It covers hairpins that are sometimes but not always present It also covers a loop that has distinct structures that are possible namely the stable GNRA or UNCG tetraloops The figure legend gives the commands that would be run to process files with modular structures The basic idea for processing modular structures in R2R is 1 to define a predicate that de termines which sequences exhibit a given substructure e g which sequences have the optional hairpin 2 extract those sequences i
33. label length Set a fixed length for a variable length backbone and do not allow the solver to modify its value label must specify the first label used in the var_backbone_range command For example given a command var_backbone_range X Y you might set multistem_junction_ solver fixed_var_backbone_length X 5 61 length is specified in units of internucleotideLen default 0 105 inches and can be any real number Bad values even negative values are allowed but will produce weird results draw_circ Tells the computer to actually draw the main circle that is the target for layout of the mul tistem junction This is useful for debugging to see what it s trying to do and might be useful in Illustrator since it might be helpful to have the circle that you can cut pieces out of draw_zero_degrees Tells the computer to draw a line in the direction of O degrees within the multistem junction solver command This is useful as a guide if your enclosing stem is directed in an unusual direction and you wish to align nucleotides flipstem 2 Flips the it stem so that it goes inside the multistem junction circle Useful for saving space align_stem_horiz st only with solver The literals s and t are stem numbers The solver will attempt to ensure that the midpoints of their base pairs that are on the multistem junction have the same X value i e are horizontally aligned Warning horizontal is defined relative to the
34. layout determined by the solver This will happen if the R2R_DUMP_FILE shell environment variable is set and this environmental variable specifies the path to the file With _solver the computer will print some additional diagnostics It will tell you the coor dinates of the center of the circle and its radius mainCircle Note these coordinates are in 63 inches and they re in a coordinate space where the left nucleotide of the enclosing stem is at the origin 0 0 and the vector from this left nucleotide to the right nucleotide is 1 0 Up is negative in the Y dimension It will also tell you the result of the objective function with the optimal variables The objective function is broken into these components e circleJoinedValue how close the total angles in the circle are to 360 degrees This should be very close to zero e alignConstraintObjective irrelevant Alignments are now specified as non linear con straints e changeInRadiusSum the main optional number My feeling is that it s best if there s as few as possible changes in the radius distance from the circle center that can arise between stems and junctions Essentially this makes the circle as smooth as possible e rightNucFakeToActualDist essentially another way of looking at circleJoinedValue e stem intersect fraction the final value between zero and one with s angle ai option e junc text col raw var backbone length the lengths chosen
35. like this using the simpler multistem_ junction_circular function However this function won t take into account var_backbone_ range regions whose lengths can be changed so it won t be a general solution e initial first point_angle number Only applicable to multistem_junction_circular_solver Specifies an angle in degrees of a nucleotide immediately 5 to the right nucleotide in the enclosing base pair The solver will use this angle as an initial value in the optimization The starting values of variables can make a big difference in whether the optimizer ends up with a good solution an acceptable but not ideal solution or a complete mess Sometimes you can get good starting values where necessary from less constrained problems that the computer has solved The computer will output the optimal first angle value after solving a problem The default is 70 e initial_var_backbone_length label length Set an initial value for the variable length backbone length to help the optimizer reach a good solution label must specify the first label used in the var_backbone_range com mand as described under fixed_var_backbone_length above length is specified in units of internucleotideLen default 0 105 inches and can be any real number However the solver will constrain it to be at least 1 The default initial length value is 2 All commands will optionally append to a file a multistem_junction_bulgey command that statically models the
36. pairbonds Commands can be conditionally performed only in skeleton mode or never in skeleton mode see Section 5 8 1 5 5 5 Entropy mode experimental feature If the line has two fields and the second field is equal to the literal string entropy R2R will draw each alignment position s entropy In this mode columns are not removed by the GC R2R_LABEL line or if the GC cons line says it s a gap Similarly var_hairpin and any var_backbone commands are ignored Rather columns are only removed if there is a dash in the special GC ENTROPY_DEL_COLS line This is to allow you to see all the columns with their entropy and also to extend to more flanking sequence The resulting diagram is drawn twice once with normal letters colors note gap columns are drawn with an circle that has a gray line rather than the usual black line and then with entropy The calculated entropy in the file by GSC weighted consensus is drawn This entropy counts the gap character as a 5 nucleotide Therefore the entropy E is guaranteed to be in the range 0 lt E lt log 5 2 3 An entropy of zero is mapped to red and log 5 is blue In between these extremes the colors are used linearly along the log scale Entropy diagrams are experimental and the code is not guaranteed to work 5 5 6 Cleavage diagrams Cleavage diagrams are specific to in line probing experiments 10 although the scheme might be applicable to other structural
37. pointing up is taken by adding 90 to the direction of the backbone at the A The two zeroes mean that no absolute positioning is used One drawing unit is equal to the distance between consecutive nucleotides along the backbone B Hypothetical Stockholm alignment to demonstrate place_explicit command contents of the file demo demo pe sto Note that the f at the end of the first and second place_explicit commands directs R2R to flip the backbone such that the 3 side of a base pair is on the left of the 5 side instead of the usual right side The second flipping command flips the orienation back to the default for the 3 tail of the hypothetical motif Synthesizing an appropriate joke based on the organisms from which the hypothetical RNA sequences were taken is left as an exercise for the reader C Annotated raw output of R2R when run on the alignment in part B Nucleotide positions that were the referent of place_explicit command i e a b and c have a blue arrow indicating the direction of the backbone at that position D Raw R2R output of the mini ykkC motif demo 22 mini ykkC sto 13 which uses a place explicit command to change the direction of the var_backbone_range just before the SD sequence 21 place_explicit command are specified in drawing units where 1 drawing unit is equal to the standard internucleotide distance between consecutive nucleotides by default 0 105 inches The direction of the nucleotide
38. sequence in the MSA and is a literal colon to the keywords DEL_COLS CLEAVAGE R2R_LABEL or R2R_XLABEL For example GR NC_003888 3 5 50 CLEAVAGE L In the rest of the sto file and in the r2r_meta file this RNA would be called NC_003888 3 5 50 L i e concatenate the hitId with the colon stuff Presumably the L here stands for your Large construct This scenario is illustrated by the yjdF motif 17 see demo 104 yjdF sto and demo 104 yjdF r2r_ meta In this case I performed in line probing experiments for the same yjdF RNA homolog but with different 3 ends You could also solve this problem using defines and draw cleavage markings explicitly with the circle_nuc command instead of using GR CLEAVAGE lines 5 6 The R2R solver cache This section applies to the solver commands for drawing multistem junctions If you do not use these commands in a sto file this section is not relevant Since the solver commands often take a long time to run I have implemented a feature to re use already calculated solutions Thus it will be slow the first time but will be quick in subsequent times unless you change the parameters of the solver command in which case R2R must compute the solution for the altered problem In other words I implemented the standard cache strategy Most likely this functionality will just work and be transparent but it is something I ve imple mented very recently If you change par
39. since it leads naturally to a callout style of pseudoknot drawing 3 8 1 A note on the representation of pseudoknotted secondary struc tures within Stockholm files R2R uses a non standard method for writing consensus secondary structures for motifs with pseu doknots To describe pseudoknots in addition to the standard GC SS_cons line there can be other lines of the form GC SS_consX for any string X typically I call them GC SS_cons_1 GC SS_cons_2 In the pseudoknot examples in this section you will see GC SS_cons_1 used GOLLD RNA 15 has 5 pseudoknots see demo exceptional GOLLD sto and uses up to GC SS_cons_5 When you use extra GC SS_cons_ lines I recommend that they each only contain only one stem possibly with bulges internal loops and not more complex structures R2R is best tested with this convention I find this representation of pseudoknots within Stockholm files to be convenient because all GC SS_cons lines are interpreted in the same way 3 8 2 Pseudoknots drawn with callouts Figure 3 16 gives an example of an RNA with a pseudoknot that is drawn using a callout style This style depicts the pseudoknot pairing in a callout Figure 3 16C that also allows R2R to shade covarying base pairs To achieve this we consider the pseudoknot as a modular structure and make 32 Figure 3 16 A pseudoknotted RNA drawn in callout style A STOCKHOLM 1 0 rope AGGCAUUUGAACCAUAUUGUGCGCCUAACA
40. standard coordinate system in which the direc tion of the left base pair of the enclosing stem is zero degrees If you are confused add the draw_zero degrees directive Warning adding alignment constraints can easily make the problem over constrained and the computer might not be able to achieve all constraints or might end up with a completely messed up layout align_stem_vert st Similar to align stem horiz but align vertically align angle angle s t Similar to align_stem_horiz but aligns at an arbitrary angle specified by angle The com mand align_stem_vert s tis equivalent to align_angle O s t while align_stem_horiz st is equivalent to align angle 90 s t If angle is the letter h then alignment is horizontal If it s v then alignment is vertical Mathematically the computer is projecting the base pairs midpoints onto the line defined by angle using a scalar projection and constrains the projected positions to be the same align_nuc_centers_angle angle list1 list2 A more general version of align angle listl and list2 each specify a set of one or more nucleotides The computer will calculate the midpoint arithmetic mean of each of these nucleotide sets and force the midpoints to be aligned The nucleotides within the sets can be any nucleotide within the multistem juction list1 and list2 are specified by a space separated list of labels that are each terminated by a dot The first thing in either list can b
41. the next 5 most stem Junction J1 identifies the junction between this next 5 most stem and the next stem to that i e the immediately 3 stem Positioning of JO is equivalent to a place_explicit command that positions X relative to Y where X is the left nucleotide of the stem on the 3 side of the junction JO and Y is the left base paired nucleotide of the stem on the 5 side of the junction JO For junctions J1 J2 etc Y is the right nucleotide of the stem on the 5 side of the junction An equivalent viewpoint is that X and Y are the base paired nucleotides immediately 3 X and 5 Y to the nucleotides within the junction This is also illustrated in Figure 3 9 The multistem_junction_bulgey command also provides alternate layouts for the bulges such as flipping the direction of the bulge and some linear layouts for the junctions It is also possible to position the 5 base paired nucleotides in all stems relative to the enclosing stem rather than to the previous stem by using J1 base in place of J1 These variations are described in Section 5 8 3 1 3 6 3 Automated multistem junction layout with direction constraints R2R provides the ability to automatically determine a layout of a multistem junction given user specified directions in which each stem must go This functionality requires that CFSQP is available see Chapter 2 This automated layout facility is refered to as the _solver commands in this manual and is
42. 0 5 is same as strategy nm strategy 1 is same as strategy r Note strategy aa constrains the problem and with solver can avoid some solutions when its applied to the enclosing stem if you apply strategy aa to the enclosing stem this constrains the center of the circle to lie on the right bisector of the line between the nucleotides of the enclosing pair while otherwise the center of the circle is not constrained strategy ai ai automatic intersect in this case Same as strategy except the number becomes a free variable in the optimization Only available for solver strategy ai Similar to strategy ai but in this case you specify the starting value before the optimization from 0 to 1 This can be useful in cases where certain starting values lead to infeasible parameters strategy ar Doesn t work well This is similar to strategy ai but formulates the problem slightly differently the free variable is the radius of the left nucleotide to the center of the circle rather than the circle intersect I thought it would lead to an easier function to optimize since the anyangle possibility should be easy to reach where with strategy ai the anyangle possibility requires going to the left or right Unfortunately strategy ar seems to perform worse e all stems anyangle Equivalent to setting strategy aa for all stems Does not work with _solver commands since there s no point e fixed_var_backbone_length
43. 0 a 50 G 6G G GecO Sy Cc CCVPACUGUCC G co Lt parr cco GS UGACAGG G A C Qa A G 092 20 30 o G go 8 s U A 908 110 l 10 G C CGGC ACCU GOO A G C 5 OOOODOCGACA OCUCGUGGACES 100 Examples of oneseq and skeleton drawings using SAM IV riboswitches 16 These examples are built from the files demo SAM IV SAM IV sto and demo SAM IV SAM IV r2r_meta A oneseq drawing of a SAM IV riboswitch from Streptomyces coelicolor annotated with cleavage data from in line probing experiments 16 The annotation of cleavage data used a GR CLEAVAGE line as explained in Section 5 5 6 Note that this is the raw output of R2R and some adjustment in Adobe Illustrator or Inkscape would be needed to position some of the nucleotide number labels in appropriate places B R2R skeleton drawing of SAM IV riboswitch consensus This layout is the same as in Figure 3 10D but at a reduced scale For oneseq drawings nucleotides are removed from the drawing using the DEL_COLS sequence specific line R2R commands beginning with GF R2R_oneseq bigfoot are only applied in oneseq mode for the sequence bigfoot e A skeleton style schematic can be drawn by adding skeleton with pairbonds or skeleton in the r2r_meta file Commands beginning if_skeleton are only applied in skeleton mode e Drawing parameters can be modified with the SetDrawingParam command This command can also be used in the r2r_meta file A
44. 2R output As stated previously the published drawing of GOLLD RNA 15 was based on the most commonly observed structural elements Creating the drawing based on the output of R2R therefore requires some assembly The drawing is generated from the demo files demo exceptional GOLLD sto and demo exceptional GOLLD r2r meta and this description refers to the PDF or SVG output gener ated when R2R is run on those files Most elements of the drawing come from the default drawing called GOLLD However the second multistem junction numbered from the 5 end is taken from the GOLLD skipbadd2 drawing The third and fourth domains are taken from the GOLLD d3classic drawing All other elements of the structure come from the default drawing 3 11 Making the demo output files To create the demos yourself cd into the demo subdirectory To create the PDF output in the output pdf directory run make all pdf Similarly to create SVG run make all svg I recom mend that you not run plain make or make all because of unlikely but possible conflicts with the solver cache file as the same file is used both for PDF and for SVG generation You might need to delete the files in output pdf or output svg in order to force the make command to re build them If you wish to really make the files from scratch also delete the files in the intermediate subdirectory and solver cache files in various subdirectories to force CFSQP to run Of course 36
45. 7 Meta Makefile This chapter explains the use of the MetamakeDemos p1 script which helps to automate the appli cation of the r2r command This chapter assumes that you know what a Makefile is and basically what the make command does It is not necessary to use the MetamakeDemos p1 script since r2r commands can be given at the UNIX command prompt directly The MetamakeDemos p1 script searches for sto files and generates appropriate r2r_meta files and a Makefile to build PDF or SVG output In order to create r2r_meta files in some cases you must add additional markup to your sto files This directs the script to add oneseq mode lines or modular sub structures etc You can run the resulting Makefile as follows To create all PDF output files make all pdf To create all SVG output files make all svg 7 1 Running the script To run the script you need to designate a directory where you will keep your sto files and create some subdirectories used by the Makefile mkdir intermediate output pdf output svg The script is normally run like this perl MetamakeDemos pl It searches the current directory and its subdirectories for sto files using the standard UNIX find command However it does not search the intermediate subdirectory It will create a new file Makefile in the current directory It will also create an appropriate r2r_meta file corresponding to each sto file it sees 7 1 1 Script options The script accepts the f
46. ABEL Or 52 GR hitld R2R_XLABEL You can also emulate this behavior more conveniently by definining special GC R2R_XLABEL_ label lines for each sequence 5 7 4 SS_cons All GC lines beginning with SS_cons are added like named labels see extra named labels above In this case for GC SS_cons the label name is SS_cons The actual string is normalized to use only the symbols lt gt or 5 7 5 Using labels and special labels Various commands use a parameter label which refers to the label of a column s The following are valid labels pos0 the literal string pos0 refers to the column 0 the 5 most position all specifies all positions allpairs specifies all positions annotated as pairing If only GC SS_cons is used it s equivalent to the union of columns matching label SS_cons lt and those matching label SS_ cons gt If other SS_cons lines are used it applies it to them too name label specifies all columns with the label label in the named label name Note the name and label are separated by a literal colon For example if you had a label e in the line GC R2R_XLABEL_fun you could refer to it as fun e Note that was a self describing bad joke The SS_cons lines act as named labels Thus SS_cons lt refers to all columns that are a left base pair in the GC SS_cons line label all columns with label label in the default label line GC R2R_L
47. ABEL This is a shortform for using the fully qualified name label format For example the label A is equivalent to the label A note the part before the colon is the empty string which is the name of the GC R2R_LABEL line The text two minus signs can be added to the end of any label name which gets the position immediately 5 to the label position Similarly is the position immediately 3 to the label position Obviously this means that labels cannot end in or The special notinpknot can only be used in the context of a SS_cons line For example given SS_cons_1 notinpknot the label specifies all positions that are heuristically assumed to not be a part of the pseudoknot that is assumed to be represented by SS_cons_1 That is all positions before the first lt between the last lt and the first gt and after the last gt This is useful for drawing pseudoknots using the define commands and not the SUBFAM functionality in the command delcol SS_cons_1 notinpknot The special label form col refers to column number col in the input alignment Column number zero 0 is the 5 most i e first left most column NOTE In most cases it is probably a mistake to use this label format because changes to the alignment that involve adding or removing columns will cause the numeric reference to become wrong However numeric labels are useful for computer scripts that generate R2R input wher
48. CCAAAGG 39 B ACUAUGU H GC SS_cons lt lt gt gt GF USE_THIS_WEIGHT_MAP A 5 B 1 and says that RNA molecule A should be weighted 5 times more than molecule B Then run the GSC weighted consensus command on this augmented Stockholm file The GSC weighted consensus command will use the specified weights whenever the GF USE_THIS_ WEIGHT _MAP tag is present 4 4 General command This section gives additional detail on the command that will likely not be relevant to most readers needs r2r GSC weighted consensus input sto output sto identity levels present levels maz non canon where e identity levels is an integer N followed by N real numbers N defines the number of different levels for classifying the degree of conservation of a nucleotide identity The real numbers define a minimum frequency from 0 to 1 for each level and must be in decreasing order e present levels is also an integer N followed by N real numbers In this case N defines the number of different levels for classifying how frequently the nucleotide is present The real numbers define minimum frequencies 0 to 1 for each level and must be in decreasing order e maz non canon is a real number specifying the maximum frequency 0 to 1 of non canonical base pairs Non canonical base pairs are nucleotide pairs that are not A U C G G C U A G U or U G If the number of non canonical base pairs in paired columns exceeds maz non cano
49. CUCGAGUACAC 0 0 GC SS_CONS lt lt lt O RAN 5 R Cee AC O c All stockhom files begin Adidas WiN P STOCKHOLM 1 0 this sequence by the STOCKHOLM 1 0 e e gap within this sequence name chimp human ACACGCGAAAC GCGC AAACGUGCACGG GAAUGUGAAAAACACCA CUCUUGAGGACCU This identifies the lineas bigfoot UUGAG UUCG CUCGUUUUCUCGAGUACAC being the secondary CEA GC SS_cons lt lt lt gt Pies gt gt 2 C An unpaired structure consensus aay Epari column pair seal Pi stockhom files end palf o column is not represented in the with drawing in part B because it is mostly gaps D The consensus line STOCKHOLM 1 0 inferred by R2R n A human ACACGCGAAA GCGCAA CAAACGUGCACGG luma is classifi none soa chimp GAAUGUGAAAAACACCA CuCUUGAGGACCU This column is classified ei me bigfoot UUGAG UUCG CUCGUUUUCUCGA as a gap by R2R eine a i GC SS_cons i carer ee S504 o cae bh dp lt e inte nnRnGnnnnR nCnCn lla en gt GC conss 1111141111041111101111111111111 conservation colors GC cov_SS cons 2 2 EEE DA EE in the diagrams _ P 0 E No mutation Compatible Covariation observed mutation E intermediate demo1 cons sto A A minimal Stockholm format alignment of a hypothetical RNA motif This is the file demo demo1 sto B The raw output of R2R when run on the alignment in part A Note that the text demol is part of R2R s raw output C Annotations of the alig
50. G Y R A The Stockholm alignment and R2R drawing commands for a hypothetical modular structure This is the file demo demo modular sto B Raw output of R2R To extract the Stockholm file for the predicate GNRA for example the command perl src SelectSubFamilyFromStockholm pl intermediate demo modular cons sto GNRA gt intermediate demo modular GNRA sto was used followed by inference of the consensus using src r2r GSC weighted consensus intermediate demo modular GNRA sto intermediate demo modular GNRA cons sto 3 0 97 0 9 0 75 4 0 97 0 9 0 75 0 5 0 1 The file intermediate demo modular GNRA cons sto was included in demo modular r2r_meta Similar commands were run for the UNCG and OPT predicates C Integration of the modular structures into a consensus diagram This drawing was created in Adobe Illustrator using the output of R2R shown in part B 31 R2R also provides Boolean operations that can combine the values of other predicates These predicates also exploit Perl and therefore inherit a Perl like syntax In these predicates the expres sion predValue GNRA for example would evaluate to true if the predicate named GNRA is true for a given sequence Two ampersands amp amp represents the Boolean and operation which eval uates to true if the two sub expressions are true So predValue GNRA amp amp predValue FOUR is true if the predicates GNRA and FOUR are both true Similarly two pipe symbols denote the
51. GC SS_cons_2 To mark base pairs a structure line must contain only left brackets P lt or or right brackets P gt or All other symbols are assumed to be single stranded Thus R2R does not use Rfam s method for marking pseudoknots I find that this method reduces special case code because all pairs are specified in the same way It also allows for nucleotides that can be in more than one base pair interaction not necessarily simultaneously This is occassionally relevant to conserved structures e g with alternate pairings and is convenient when the exact structure is unknown R2R does not work with consensus structures in Rfam format Supporting such a format au tomatically would require some rethinking of structure related commands such as ignore ss except for pairs 5 3 R2R drawing units R2R divides an RNA structure into units and positions each unit as a block Units are 1 a single stranded region including terminal loops 2 both sides of an internal loop or 3 a stem Some commands operating on labels will have no effect if the label corresponds to a position in the middle of a unit For example to use the turn_stem_at_internal command you must specify the 5 most nucleotide in the internal loop although in this command the 3 most nucleotide is also permitted place explicit commands that reference the midding of a drawing unit will break it in
52. If the value is negative then the computer will automatically set the width based on the size of nucleotide symbols i e the font size default 1 i e set automatically alongBackboneStyle permissible styles are integers default 0 as follows x 0 shade nucleotides with circles join consecutive base pairs with line but keep endpoints rounded This tends to look good with shade_along_backbone where nucleotides are shaded It is the default style x 1 shade nucleotides with an appropriately thick line that extends half way to the previous nucleotide and half way to the next This looks better for skeleton drawings where you want to change colors otherwise lone circles look weird when they re small and not shading a nucleotide alongBackboneMidpointGapLength If alongBackboneStyle is set to 1 this parameter allows for gaps between the shading of consecutive nucleotides By default differently colored consecutive nucleotides will have the two different colored shading paths joined at the midpoint between the two nucleotides However a gap of length alongBackboneMidpointGapLex can be left between these two paths so that they don t touch There are two technical issues that complicate this feature First the length for the gap is the length along the backbone path between the two nucleotides not the straight line length Thus if the two consecutive nucleotides lie on a circle meaning their backbone path is an arc the
53. Largely the same as var_backbone_range except that I think it draws a nicer circle in the special case where you re replacing the whole terminal loop 5 8 5 3 Variable length backbone var backbone range firstLabel lastLabel text Replace the nucleotides from firstLabel to lastLabel with a variable length backbone The nu cleotides within the firstLabel and lastLabel columns are included within the variablen length region All nucleotides within the variable length region should be single stranded or R2R will report an error that base pairs are broken However you can use the ignore_ss command to remove pairing that complicates the use of the var_backbone_range command This strategy is demonstrated in the file demo 104 SAM I IV variant sto The variable length region will be drawn with a thick black line The line will be labeled with text If text contains the special string ntrange then the occurrence of ntrange is replaced with the text x y nt where x y is the range of the number of nucleotides within the variablen length region If text is not given then it simply gives the range x y nt This is the functionality I normally use You can include the text An where is a single space and this will put the remaining text on another line Note if you re replacing an entire terminal loop use var_term_loop as it usually looks nicer Note after applying a var backbone range command the lastLabel will no longer work
54. NA ALignment Editor in Emacs Bioinformatics 21 257 9 2005 C Lawrence J L Zhou and A L Tits User s guide for CFSQP version 2 5 A C code for solving large scale constrained nonlinear minimax optimization problems generating iterates satisfying all inequality constraints Technical report Institute for Systems Research University of Maryland College Park 1997 TR 94 16r1 E R Lee J L Baker Z Weinberg N Sudarsan and R R Breaker An allosteric self splicing ribozyme triggered by a bacterial second messenger Science in press J Perreault Z Weinberg A Roth O Popescu P Chartrand G Ferbeyre and R R Breaker Identification of hammerhead ribozymes in all domains of life reveals novel structural variations PLoS Comput Biol 7 e1002031 2011 E Poiata M M Meyer T D Ames and R R Breaker A variant riboswitch aptamer class for S adenosylmethionine common in marine bacteria RNA 15 2046 2056 2009 E E Regulski R H Moy Z Weinberg J E Barrick Z Yao W L Ruzzo and R R Breaker A widespread riboswitch candidate that controls bacterial genes involved in molybdenum co factor and tungsten cofactor metabolism Mol Microbiol 68 4 918 932 2008 G A Soukup and R R Breaker Relationship between internucleotide linkage geometry and the stability of RNA RNA 5 1308 25 1999 N Sudarsan E Lee Z Weinberg R Moy J Kim K Link and R Breaker Riboswitches in eubacteria sense t
55. OuterLabel and rightOuterLabel must pair with each other although this restriction could be relaxed with improvements in the code 5 8 6 Annotation 5 8 6 1 Tick labels like the nucleotide numbering in cleavage diagrams tick_label label text Annotate the nucleotide identified by label with text The text is connected to the nucleotide with a tick mark i e a short line Within text you can include the text An where J is a single space and this will put the remaining text on another line Thus the command tick_label A one line n two line n red line n blue line will take 4 lines tick_position_label_names Useful for debugging and helping you work out where things are Labels nucleotides with all valid labels except for the SS_cons implicit labels like SS_cons lt tick_label_regular_numbering start skip zero Uses the tick style labeling to label nucleotides with their number The nucleotide position start is labeled as is start skip start 2 skip When one sequence is printed like to show ligand dependent changes in cleavage there is implicitly start 0 skip 10 Note that the first 5 most nucleotide is number 1 so in the oneseq case the 5 most nucleotide is not labeled and the first labeled nucleotide from the 5 end is nucleotide number 10 If the optional zero is put at the end of the line then the numbers are zero based This is mainly useful if you re using the labels to aid in debugging the C
56. R2R version 1 0 3 user manual Software to speed the depiction of aesthetic consensus RNA secondary structures Zasha Weinberg August 15 2012 0 6 nt pseudo knot G C A U C R A U Copyright c 2009 2012 by Zasha Weinberg All rights reserved This software package including source code and this manual is distributed freely but without any warranty under the terms of the GNU General Public License as published by the Free Software Foundation This program is distributed in the hope that it will be useful but WITHOUT ANY WARRANTY without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE For a copy of the full text of the GNU General Public License see www gnu org licenses This package includes the following material from other sources e The Squid library from Infernal version 0 7 by Sean Eddy e Free code that is part of the CFSQP package 5 distributed by AEM Design e Previously published layouts of RNAs and R2R markup to generate those layouts 13 17 15 16 12 9 11 6 1 2 Introduction 1 1 What does this software do a a a a aa 1 2 What does this software not do lt lt lt 2 Lo Cd ida a e A A a a o e a e E E AI 1 5 Changes rom previous Versione ci a AAA A G Installation 2 1 Platiorms on which R2R 18 known to work o e scu exar BR eo 2 2 Decide whether to install CESOP or not eco c esser oee reses
57. The subset predicates allow you to do this They also allow you to draw structures that appear only in some RNA sequences like an optional stem 6 2 1 Command line To run SelectSubFamilyFromStockholm pl perl SelectSubFamilyFromStockholm pl input sto file predicate name gt output sto file 6 2 2 How to define predicates You define the predicate within the sto file To define a column or columns in the alignment use the following format GC SUBFAM_LABEL_columns name 77 Then annotate each column where an X means that column is referenced by the predicate and a dot means it is not There are two types of predicates regex or perl 6 2 2 1 Regex predicates Regex predicates match all seqs whose selected columns match a given regular expression The format is GF SUBFAM_REGEX_PRED predicate name columns name regex where predicate name is the predicate you re defining columns name designates the columns referenced which you defined using GC SUBFAM_LABEL_columns name and regex is a valid regular expression in Perl For example GC SUBFAM LABEL GNRA XXXXXXX don t take this literally see the GEMM sto file from Sudarsan et al for specifics GF SUBFAM_REGEX_PRED GYRA GNRA G CU AG A In this example we define a set of columns GNRA which is the 7 nucleotides positions columns in the GNRA tetraloop of GEMM RNAs Sometimes there are more nucleo
58. UC GCCAAAGCACAA speed GGGUAUUUGAACUGUAUUAUGCACCCAGCAUAAUGUGGAACCAUAA topology AGGUAUUUGAACCGUAUUGUGCACCUAGCAUGA GUUAAAGCACAA GC SS_cons NN E NN GC SS CONS L amnesia Ed gt gt gt gt H GC R2R LABEL ante Shaican arses ere ei oneretinl aa veas aero E Dieraiiane snares GC SUBFAM_pknot_R2R_LABEL sos GF R2R ignore_ss_except_for_pairs _1 outline GF R2R var_backbone_range 1 2 H GF R2R place explicit 3 3 45 10020 90 GF SUBFAM_PERL_PRED pknot return 1 H GF SUBFAM_pknot_R2R subst_ss _1 primary GF SUBFAM_pknot_R2R no5 H GF SUBFAM_pknot_R2R outline_nuc all H GF SUBFAM_pknot_R2R set _dir pos0 90 f B C A demo pknot callout G U YR U AC Ua U A U A Ay R G U A G C U R Y G C U U A R Y Ay R G C A G C A 5 PTYARCAU crer 1 3 nt demo pknot callout pknot subfam_weight 1 REM U A G C c 0 GYR u The demo motif U A U pseudoknot A R Y U A G C c 0 gt gt 0 gt lt D cYe A The Stockholm file demo demo pknot callout sto The structure of this hypothetical RNA is in spired by that of SAM V riboswitches 8 but is simplified for this example B The raw output of R2R when run on demo demo pknot callout r2r_meta demo demo pknot callout sto C Assembled drawing To create this drawing the components of part B were combined using Adobe Illustrator 33 R2R draw it separately Modular structures were explained in Section 3 7 Their applic
59. ameters and the solver does not re run try deleting the Solver cache file see below for where it is Note that even parameters not part of the solver command can force a solver re run For example if you add a variablen length region within the mul tistem junction or change certain drawing parameters like internucleotideLen this indirectly changes the problem and the solver must solve the new problem Stored cached solutions are associated with each r2r_meta file you use as input If you process a file called something r2r_meta and there is a solver command used in it R2R will create a file called something r2r_meta solver cache This solver cache file contains cached solutions The solver cache will not be created if the given r2r_meta file does not use any solver commands I have attempted to make the solver cache files platform independent so that you can create a solver cache file on one system and use it on a different one However this functionality is 51 not well tested In principle you should be able to build the demo files without CFSQP because of the solver cache files I ve provided 5 7 Labels Labels allow you to identify a column or columns in the alignment by a name Defining column s by names is somewhat more convenient that using the column number and also means that the R2R markup remains valid even if you add or remove columns Therefore R2R only supports referencing columns by name Tip
60. ary structures 43 Bad RoR drawing DIS cocido sora das eee RES oe eS 43 24A Datatypes n EE 44 Dal WR os ndaid a ea eh A e RASH a a iodo de a 5 44 DAS AOS a ee e ee A eR A 44 A A A kG OE d A 44 5 4 4 Measurements length width size o ooo eee 44 5A Cols AI GEA amp Ch GOS CRE ee Fe 44 DG NRO ooe su ioraa ee GA AT Be eS Sw bee ds 45 AM AE 45 59 2 ds IE EASE EE ES 45 da TIMO e e Daria Bee a A A A eG 49 A A 2 4 4 e ee wk Bw eH Ee ee ee eee we eS 50 Gos Ee DS ony EE LAA AAA 50 56 0 Cleavage diagrams lt lt ecos 6 dau droa kordas sato 50 5 5 6 1 Multiple drawings of the same RNA molecule 51 mo The ER SONETOS oe erani Ga ean AAA AA 51 A AAA 52 orl Man labels e cme eh em ba eH SES A A A 52 ara Extra nomed label lines s s esera ede sr ed eee A 52 5 1 3 Sequencespecillo optional os c sa i084 edie REE rikota sa 52 pte IIA a er REA EEN SES A A eS Eee 53 5 7 5 Using labels and special labels a aa RRA a RR 53 o E e e a e EE 54 D81 Conditional Commands cs sa bga do ekaa ae BK ia k Aa k 54 ALL Dehnesymbols lt pess see e e a A a AA 54 5 8 1 2 Commands that apply to only a consensus or single molecule drawing 55 2 9 9 5 10 5 8 1 3 Other kinds of conditional commands 2 55 G82 Turning and posiioning ssc e ces owe bw EERE ERGO AA 55 ahal IIA e iasa AN EY ORES ER OES 55 5 8 2 2 Laying out arbitrary units like a bulge 56 5 8 2 3 Layout single stranded loo
61. at label is the angle at relative label rotated by next angle degrees An optional f at the end of the command directs R2R to flip the stem Figure 3 7B C Flipping is analogous to applying a twist to the RNA Where R2R normally positions the 3 nucleotide of a base pair to the right of its partner flipped regions will cause the 3 nucleotide to be on the left Flipping is used only sometimes but Figure 3 7C shows a typical example where it is useful to avoid crossing stems A real world example of this usage is the IMES 4 motif 15 see demo exceptional IMES 4 sto Section 5 8 2 6 explains the command in more detail More advanced usages of place_explicit are necessary for the in line layout of complex pseudoknots Section 3 8 3 I explain more about place_explicit commands and their semantics in that section You can use set the showPlaceExplicit variable to true in order to see the place_explicit commands in the context of the structure This is demonstrated in the same section Section 3 8 3 3 6 Multistem junctions and turning internal loops Multistem junctions are a structural element within a RNA secondary structure in which a loop includes more than two base pairs e g see Figure 3 8 This section can be skipped if you are drawing RNAs that don t have multistem junctions However the material in this section is also sometimes useful for turning the direction of the stem at internal loops see Figure 3 14 for an examp
62. ation to pseudoknots is somewhat simpler since all sequences presumably have the pseudoknot so there is no need to define any regular expressions The line GF SUBFAM PERL _PRED pknot return 1 e g see Figure 3 16A defines the pknot modular structure as applying to all sequences R2R commands and GC lines beginning with SUBFAM_pknot_ then define commands that should be run to generate the pknot structure The actual drawing is initiated by a line in the r2r_meta file and running extra commands to create the pseudoknot view of the RNA structure as in Section 3 7 It is also possible to draw callout style pseudoknotted structures without the use of the modular structure logic but instead using the define and ifdef commands to draw the main RNA and separately draw the pseudoknot The associated files demo demo pknot callout ifdef sto and demo demo pknot callout ifdef r2r_meta demonstrate how to do this Once R2R has drawn the pseudoknot and the main structure these need to be assembled in a drawing program like Adobe Illustrator or Inkscape One style of this assembly is shown in Figure 3 166 Note R2R will attempt to automatically join the lines or arcs that comprise an outline corre sponding to a pseudoknot However sometimes its heuristics will fail and it might be missing some lines 1 e some nucleotides will not be outlined In this case try disabling the automatic joining behavior GF SetDrawingParam outlineAutoJoin fal
63. ault 0 5pt outlineNucColor color used Default rgb 92 92 92 circleRadiusToSmoothDirectionChange when two straight lines in the outline meet the direction change is smoothed using a circular arc This parameter is the radius of the arc Default 0 025in outlineAutoJoin if true R2R will attempt to join lines and arcs that are part of the outline for both outlines generated implicitly by pseudoknots and by outlines generated explicitly with the commands outline_nuc or inline nuc Sometimes R2R s heuristics for drawing outlines fail and so it is better to doing the joining yourself Default true e Circling nucleotide also used for cleavage diagrams in oneseq mode cleavageIndicatorRadius radius of the circle Note R2R will refuse to allow this to be greater than half the length of internucleotideLen since then circles would overlap Default 3 75pt cleavageIndicatorPenWidth width of line surrounding the circle Default 0 5pt e Drawing base pairs bonds pairBondLen the length of the line connecting Watson Crick base pairs Default 0 054in pairBondWidth the width of the line connecting Watson Crick base pairs Default 0 02in pairBondGURadius radius of the filled circle for representing G U wobble pairs only used in oneseq mode Default 0 02in pairBondNonCanonRadius radius of the un filled circle for representing non canonical base pairs only used in oneseq mode Default 0 01in
64. best layouts found by a person Therefore R2R assumes that a user will optimize the layout and tell R2R what to do e R2R does not have a graphical user interface As noted above R2R is aimed at bioinfor maticians Successful use of R2R will likely require some general familiarity with the UNIX command line and comfort with command driven programs Because of the lack of a graphical user interface R2R might have a significant learning curve So if you just want to draw one RNA it might be most efficient to go straight to Adobe Illustrator Inkscape or CorelDRAW R2R is not designed to produce drawings that illustrate many elements of tertiary structure or whose layouts are based on atomic resolution 3 D structures R2R is only intended to create more abstract layouts of secondary structure that reflect Watson Crick base pairing 5 e R2R is not a fully general drawing program or even a fully general RNA drawing program R2R should be used in combination with general purpose drawing programs like Adobe Illus trator or Inkscape 1 3 Credit If you use R2R please cite the paper 14 If you distribute R2R or derivatives of it please continue to credit Infernal as on the first page of this manual 1 4 Licensing The files in the main directory documentation the src and the demo subdirectories are copyright 2009 2010 by Zasha Weinberg except as noted The entire package is distributed freely but without any warranty whatsoever und
65. between the distances of the stem nucleotides and the radius of the circle perfect smoothness means that the stem nucleotides will lie exactly on the circle On the other hand the commands multistem_junction_bulgecircley_solver multistem_ junction bulgecircleynormals solver and multistem junction bulgecircley stemnormals solver require that junctions have to lie on some circle but not necessarily all on the same circle The computer then tries to find the junctions that fit a common circle as closely as possible There are two specific ways to do this which actually seem to give similar results With multistem_ junction_bulgecircley_solver the computer measures the distance between points on the junc tion and the common circle The points are evenly spaced and so the computer is using a finite approximation to the integral which would be hard to compute The computer minimizes the mean square difference between the distances With multistem_junction_bulgecircleynormals_ solver the computer attempts to minimize the differences between the normals of the common circle curve to the normals in the bulge s circle at the given points The normal is the vector perpendicular to the circle at a given point The computer measures differences between normals by the square of their dot products multistem junction bulgecircley stemnormals_solver only checks for the deviations of the normals at the nucleotides in a stem i e at the ends of a bulge
66. blueberry AAACAUCCCCAU UGG smurt AAGCAACCCCUU CGG smurf AAGCAACCCCUU CGG GC SS_cons lt lt lt gt gt gt GC SS_cons ni GC R2R_LABEL Suks E F STOCKHOLM 1 0 G fo CEM det lemur AAACUACCCCUAUUGG demo breakpair fix4 lemur AAACUACCCCUAUUGG blueberry AAACAUCCCCAU UGG cc blueberry AAACAUCCCCAU UGG enue AAGCAACCCCUU CGG ie smurf AAGCAACCCCUU CGG GC SS_CONS lt lt lt lt gt gt gt gt 0 a PPP PKs GC R2R LABEL p Do R Y GC R2R_LABEL p p H GF R2R keep p 5 AA GG H GF R2R depair p A input Stockholm file demo demo breakpair sto err Column numbers are labeled with the left most column having the number zero This is the numbering system used by the text editor pro B output of R2R with an error since the base pair is not well conserved remove it from the SS_cons line gram Emacs The referenced column numbers are circled in blue C one resolution You should only do this if you truly believe on reflection that the pairing is not biological file demo demo breakpair fix1 sto D typical resolution the base pair is not that well conserved anyway so don t show its nucleotides in the consensus diagram at all However retain the pairing in the align ment since the pair is sometimes formed The nucleotides in the pairing are removed from the drawing by putting dashes in the R2R_LABEL line file demo demo breakpair fix2
67. bol in the GC R2R_LABEL line that causes the column to be deleted from R2R s drawing Also it is important to note that a single space character separates all elements of R2R commands This text is identical to the file demo demo1 ii sto B Raw output of R2R with annotation in blue showing what elements of the drawing were the result of the R2R commands in part A e R2R commands refer to specific alignment columns using labels in the GC R2R_LABEL line It is also possible to expand the number of labels beyond simply one symbol see Section 5 7 e Columns can be deleted by putting dashes in the GC R2R_LABEL line e R2R has commands to represent variable length regions of different sorts For more informa tion on R2R commands related to variable length regions see Section 5 8 5 e Specific positions can be labeled with a tick mark using the tick_label command 3 3 3 Other kinds of drawings single molecule and skeleton diagrams Figure 3 3 demonstrates two additional types of R2R drawings The r2r_meta file is needed to specify both kinds of alternate drawing The first is drawing of a single molecule whose sequence is present in the alignment Every tenth nucleotide is numbered by default These kind drawings are called oneseq drawings because they depict one sequence A real world example of this kind of drawing is given in Figure 3 4A The second kind of drawing is a skeleton drawing which is an outline of the
68. bulges label When label is the beginning left nucleotide of an internal loop within a hairpin splits each side of the hairpin into bulges so that they can be manipulated independently This is also important if you want to use the command bulge label 5 8 4 4 Ignore pseudoknots entirely ignore_ss ssname Entirely ignore the given secondary structure line For example ignore ss 1 will ignore the pairings in the line GC SS_cons_1 ssname primary is a special name that corresponds to SS_cons i e the primary secondary structure This is useful because Emacs RALEE gets upset with lines that end in whitespace Thus ignore_ss primary will ignore the pairings in the line GC SS_cons 5 8 4 5 Ignoring pseudoknots for the purposes of layout ignore_ss_except_for_pairs ssname detail Do not position nucleotides at all based on the given secondary structure consensus line The secondary structure line is specified in the same way as with the ignore_ss command see imme diately above If detail is outline then outline the nucleotides involved in base pairing This is useful for callout layouts of pseudoknots If detail is outline_only_bp this is similar to outline but it only outlines the nucleotides that are really involved in a basepair By contrast outline will attempt to also outline bulges If detail is ignore then don t do anything with the base pairs 5 8 4 6 subst_ss subst_ss replace ss into ss deletes t
69. but you can refer to the region using the label firstLabel Note in single stranded and straight regions the length of a var_backbone is determined by the size of the text that it s labeled with But in circular regions bulges or other loops the var_backbone length is fixed by default 3 regardless of the size of the text 66 var_backbone_range_if 5prime_nuc_exists firstLabel lastLabel text Same as var backbone range but only count sequences where there is at least one nucleotide that is 5 to the variable backbone This is to account for sequences that are truncated like environmental sequences that shouldn t be counted var_backbone_range_size_fake_nucs numNucs firstLabel lastLabel text Same as var backbone range but size the backbone as if it contained numNucs nucs This is silently ignored if the backbone is in a straight region i e not an internal loop terminal loop or bulge numNucs can be any positive number not just integers but it will usually look bad if it s less than 1 5 8 5 4 var_stem var_stem leftOuterLabel leftInnerLabel rightInnerLabel rightOuterLabel Replaces a stem region with a variablen length stem The original stem region is allowed to contain bulges internal loops but not terminal loops The region is defined by the closed intervals leftOuterLabel leftInnerLabel and rightInnerLabel rightOuterLabel which are assumed to pair to each other left pairs with right It is required that left
70. cw ter tss La Make grd party 0 o ees es peos eeose A A ea as Dial irnal e os eh BSS ea ee a aga ea dad dl ESO PO ocg co ee ee gie ee g ie A AA De Make RIR cards rr A a a 45 24 1 Optonsliy edit Maken lt o cosc 25 4 fe ew ceda h 2 4 1 1 Optionally enable CFSQP ccc saacana dani 24444 4405 2 4 1 2 Optionally choose an SVG Wt 2 4 2 2 ee ee eee ees A oe oe ee ae Mee ee eee BEG eS ee ES 2 5 Adobe Reader glitch 2k ke ee eRe RR wR ba Ew SE oo 2 6 Alignment editor software ee Tutorial a guide to R2R with examples 31 About the AE s a RN Re RA RR A 3 1 1 Demo output is already provided o 3 1 2 Drawing programs Adobe Illustrator or Inkscape 312 1 More technical informatio 22 4454 4448440546 4 a 3 1 3 Where the demo files come from 2 2 0 ee ee ee ee 3 2 Example legend to help you with finished drawings 3 3 A hypothetical motif to illustrate the basics 0 00 0 0 ee eee OL Delaplit output ol ROR occiso a E ka 3 3 3 2 Some typical customizations for consensus diagrams 3 3 3 Other kinds of drawings single molecule and skeleton diagrams 3 4 Common error One or more pairs is getting broken by one side getting deleted 35 The place explicit command icons RRA 3 6 Multistem junctions and turning internal loops a a 3 6 1 Simple circular layout cocos cee e ee RE 3 6 2 Manual layout of multistem junctions
71. d all nucleotides with label The optional width directive specifies the thickness of the line forming the box By default it s the same as for the cleavage diagram 5 8 6 7 Outlining inlining nucleotides outline _nuc label color Draws outline around the outside of all nucleotides with label This is the same look as for drawing pseudoknots color is currently ignored inline_nuc label color Same as outline_nuc but draw the line on the other side Warning R2R will attempt to automatically join the lines or arcs that comprise an outline However sometimes its heuristics will fail and it might be missing some lines In this case try disabling the automatic joining behavior SetDrawingParam outlineAutoJoin false 68 5 8 6 8 Boxing a set of nucleotides box label label X Y text Draws a bounding box in gray around the nucleotides identified by label The box is labeled with text in the direction specified by X Y X and Y are both either 1 0 or 1 For example box_label L 1 O this is a box would box the nucleotides identified by the label L and put the text this is a box immediately to the right of the box 5 8 6 9 Shading nucleotides along the backbone shade_along backbone label label2 color Draws a thick line along the backbone defined by the nucleotides belonging to any of the labels in the list The last parameter is always the color of the shading The line is thick enough to fully shade the nucleotide
72. demo Additional file 4 svg for SVG format that can be imported into Inkscape 3 3 A hypothetical motif to illustrate the basics Now the actual tutorial begins 3 3 1 Default output of R2R Figure 3 1 shows a simple alignment and R2R s drawing of the consensus of the alignment Elements of the Stockholm format alignment file are illustrated The drawing was made by running the following commands within the demo directory Actually I really just ran make which reads the Makefile in that directory to run these commands If you re familiar with the UNIX make command and you re going to make many drawings you might want to read Chapter 7 after reading this tutorial First in order to draw the consensus structure we must first run a command to calculate it We calculate the alignment s consensus Figure 3 1D using the following command line note the following command should be one line but was split into two lines for typesetting commandprompt src r2r GSC weighted consensus demol sto continued intermediate demol cons sto 3 0 97 0 9 0 75 4 0 97 0 9 0 75 0 5 0 1 All commands in this section should be run within the demo directory that is part of the R2R distribution The general form of this command is 13 Figure 3 1 demol A simple alignment and R2R drawing A B STOCKHOLM 1 0 demo1 human ACACGCGAAA GCGCAA CAAACGUGCACGG 00 chimp GAAUGUGAAAAACACCA CUCUUGAGGACCU e bigfoot UUGAG UUCG CUCGUUUU
73. e Note R2R is internally hardcoded with font geometry that is appropriate to Helvetica Arial Bitstream Vera or DejaVu fonts These fonts have similar sizes to each other If you use fonts whose symbols have significantly different sizes nucleotides will not be positioned correctly 2 4 2 Run make Finally run go into the src directory and run make Note the first file ParseOneStockholm cpp might take a surprising amount of time to compile This is normal The file is just very large If everything works there will be an executable file named r2r within the src directory 2 5 Adobe Reader glitch This section describes a potential problem when viewing PDF output with Adobe Reader With re cent versions of Adobe Reader and certain computer configurations circles and arcs appear jagged more like polygons To prevent this while running Adobe Reader select Preferences under the Edit menu Within the category Page Display uncheck Use 2D graphics acceleration and make sure that Smooth line art is checked 2 6 Alignment editor software R2R s input is based upon multiple sequence alignments that are stored in Stockholm format files Although these files can be edited in any text editor some programs are customized for them One example is RALEE 4 which is a set of macros for the Emacs text editor 10 Chapter 3 Tutorial a guide to R2R with examples This chapter gives a practical introducti
74. e This section defines the mapping from positions in the drawing to columns in the original input alignment The 5 most position in the drawing is numbered zero and the 5 most column in the input alignment is also numbered zero Each data line in the section has two fields A position index within the drawing The corresponding column in the original input alignment 75 e Header line layout num data lines The fields in the data lines are as follows Position index of nucleotide within the drawing The nucleotide letter A C G X coordinate Unit length is inches which is R2R s internal unit Y coordinate Is backbone flipped 180 degrees at this position like mirror image e g in commands like bulge flip 1 Yes 0 No Does the backbone from this position to the next position i 1 go along a circle 1 Yes along a circle 0 No along a straight line Note the radius of the circle is equal to the distance from the center of the circle to the position of the nucleotide If the backbone is linear i e not a circle at this position then the orientation angle of backbone at this position in degrees Otherwise undefined If backbone along a circle at this position X coordinate of the circle s center Otherwise undefined If backbone along a circle Y coordinate of circle s center If backbone along a circle does the next posi
75. e ifdefeq commands can apply to a single line or to multiple lines exactly like ifdef commands ifdefneq name value opposite of ifdefeg ifdefneg commands can apply to a single line or to multiple lines exactly like ifdef commands else for a multi line ifdef or ifndef reverses the test endif terminates a multi line ifdef or ifndef The following names are predefined based on drawing options selected oneseq if we re in oneseq mode then the oneseq symbol is the hit d that identifies the sequence Otherwise if we re drawing a consensus the default then oneseq is not defined skeleton true if we re in skeleton mode i e if either skeleton or skeleton with pairbonds was specified in the r2r_meta input file skeleton with pairbonds if skeleton was specified in the r2r_meta command file then this define symbol has the value false Otherwise if skeleton with pairbonds was speci fied it has the value true Otherwise it is not defined cfsqp defined only if the CFSQP solver is available entropy defined only if entropy mode is used 54 5 8 1 2 Commands that apply to only a consensus or single molecule drawing Commands given by GF R2R_allseq commands or GF R2R_consensus commands will only be interpreted if we re in drawing the consensus motif mode Commands given by GF R2R_oneseq hitld commands will only be interpreted in oneseq mode for the given hitld Setting oneseq or consens
76. e basal pair of the next stem relative to the left nucleotide of the enclosing pair from the enclosing stem For J1 J2 we position the left nucleotide of the basal pair of a stem relative to the right nucleotide of the basal pair of the previous stem This makes a bit of sense if you think of going around clockwise or equivalently it s the closest column numbers in the alignment e J basei anglePlace x1 y1 x2 y2 angleMove This is the same as the Ji command just described except that the placement is always relative to the left nucleotide of the enclosing stem In other words both JO and J base0 place relative to the left nucleotide of the enclosing stem but J base1 also places relativive to this enclosing nucleotide whereas J1 places relative to the right nucleotide of the enclosing stem positioned by JO or J base0 This function is mainly used by R2R itself so that R2R can give static layout commands that are equivalent to computationally expensive solved layouts e bi or bf or bst or bpei place explicit stuff or bss or btriangle corner or btrianglef corner Here the b stands for bulge It should arguably be j for junction but that is taken for positioning the stems surrounding the junctions The bulges junctions are numbered clockwise starting at i 0 bi lays out like a bulge which is the default anyway bfi is a flipped bulge i e the bulge goes the other way bsi lays the junction out along a stra
77. e circle center in which case the center of the 62 circle used in optimizing the multistem junction will be used circle center must be first in the list if it is used See demo THF THF sto for an example of its use If angle is the letter h then alignment is horizontal If it s v then alignment is vertical e try_harder Try harder to solve the problem by trying different initial values of certain variables and seeing which leads to the best solution The variables modified are initial_radius and initial_first_point_angle see below For multistem_junction_circular_solver I have encoded a few different values that have worked in some circumstances The list is of course not complete For the _ bulgecircley_ commands I have not come upon any cases with a strong need for alternate values Therefore try_harder has no effect on these commands e initial_radius number Specifies the circle radius in inches that the solver will use as an initial value in its opti mization The starting values of variables can make a big difference in whether the optimizer ends up with a good solution an acceptable but not ideal solution or a complete mess Some times you can get good starting values where necessary from less constrained problems that the computer has solved The computer will output the optimal circle radius after solving a problem The default is 0 459181 Note I have considered trying to set initialization values
78. e the deviation of the straight lines from an ideal connector is not usually very visible at the scale at which diagrams are drawn If anchor points a and a are associated with lines we calculate their intersection If the intersection is equidistant from the positions of a and ai 1 then we join a to a 1 using an arc that runs through a and a The radius of the arc is the distance between a or equivalently a 1 and the intersection point This case frequently arises in R2R s default layout of stems e g at the 5 end of the RNA in Figure 3 3 or when the turn_ss command or equivalent place_explicit command is used If the intersection point is not equidistant from the two anchor points we declare a join error In this case it is likely possible to extend one of the lines associated with a or a 1 in order to lead to an equidistant intersection but R2R does not implement this logic This functionality is implemented in RnaDrawer cpp 87 Bibliography E 10 11 12 R Durbin S R Eddy A Krogh and G Mitchison Biological Sequence Analysis Probabilistic models of proteins and nucleic acids Cambridge 1998 S R Eddy Infernal User s Guide 2003 ftp ftp genetics wustl edu pub eddy software infernal Userguide pdf M Gerstein E L L Sonnhammer and C Chothia Volume changes in protein evolution Journal of Molecular Biology 236 4 1067 78 1994 S Griffiths Jones RALEE R
79. e the script would simply regenerate the input upon changes to the alignment 93 5 8 R2R commands Commands are given with the text GF R2R commands All parameters are space separated with exactly one space character Yes I have not imple mented a robust parser 5 8 1 Conditional commands 5 8 1 1 Define symbols R2R implements a simple system for conditional processing that is inspired by but much simpler than the C preprocessor Symbols can be defined and R2R commands can be performed only if a given symbol is defined or not defined or if the symbol is equal or not equal to a specific value The other conditional processing commands given in later sections are supported for backwards compatibility with previous structures I have drawn but are interpreted in the context of defines The following commands operate on defined symbols define name value sets the symbol name to the given value ifdef name true if name has been defined ifdef commands can apply to a single R2R command or two multiple lines of R2R commands If the name value is at the end of the line the ifdef command will apply to subsequent commands until the next endif command If there is text beyond the name value then the ifdef command will apply only to that line ifndef name opposite of ifdef ifndef commands can apply to a single line or to multiple lines exactly like ifdef commands ifdefeq name value true if name has the value of valu
80. e to speed the depiction of aesthetic consensus RNA secondary structures 14 Briefly this software is designed to speed the drawing of RNA secondary structure consensus diagrams which show the conserved features within a set of related RNAs The software also supports drawing of single RNA molecules although this is not the emphasis To make RNA drawings many biologists use general purpose software such as Adobe Illustrator at great cost in time and risk of errors However this strategy produces the highest quality of drawings R2R is designed to allow the user to achieve this highest quality drawing while taking much less time than the manual solution Because of this goal R2R imposes more work on the user than highly automated solutions and the current version of R2R is aimed at bioinformaticians or biologists with some familiarity with UNIX like command line tools I have used R2R to draw over 100 RNA consensus diagrams These drawings are made available as demo files see Chapter 3 1 2 What does this software not do R2R has some important limitations e R2R does not fully automate determination of a layout Although its default layouts will get you closer to an ideal layout and it has functions to assist in determining layouts of multistem junctions R2R does not solve the problem of determining an overall layout No currently available computer algorithm can determine an ideal layout that is comparable in quality to the
81. ed in a multi processor make session because of possible interactions between the solver Disabled by the noal1 flag all pdf generate all PDF output files all svg generate all SVG output files clean remove intermediate and output files but keep the r2r_meta files and Makefile that were created by the script clear solver cache remove cache files used by the CFSQP solver solver pdf generate only PDF output files that use the CFSQP solver and maybe a few that don t but fool the script Used for debugging solver svg analogous to solver pdf but for SVG output concat pdf generate all pdf and concatenate the output files together so you can look at them quickly Mainly just useful for debugging Requires Ghostscript s gs command which is not distributed with R2R solver pdf same as concat pdf but only concatenate the PDF output specified by the solver pdf target 81 7 2 Additional markup for sto files The script creates an r2r_meta file for each sto input file The r2r_meta file will always include the sto file in R2R s default consensus mode but can include other types of runs 7 2 1 Implicit rules If you use a GR DEL_COLS tag for any sequence in the alignment the script will draw that sequence in oneseq mode adding a line to the r2r_meta file If you use a GF SUBFAM_PERL_PRED or GF SUBFAM_REGEX_PRED directive the script will in stantiate the subfamily alignment using Selec
82. ed nucleotide frequencies were then calculated at each position in the multiple sequence alignment To classify base pairs as covarying the weighted frequency of Watson Crick or G U pairs was calculated However aligned sequences in which both nucleotides were missing or where the iden tity of either nucleotide was uncertain e g was N signifying any of the four bases were discarded Classification as a covarying position was made if two sequences had Watson Crick or G U pairs that differ at both positions amongst sequences that carry the motif If only one position differed the occurrence was classified as a compatible mutation However if the frequency of non Watson Crick or G U pairs was more than 5 we did not annotate these positions as covarying or as compatible mutations Note it is now possible to weight sequences using methods other than the GSC algorithm see below 4 1 Recommended command The standard command line is r2r GSC weighted consensus input sto output sto 3 0 97 0 9 0 75 4 0 97 0 9 0 75 0 5 0 1 where input sto is the name of a Stockholm format file containing your alignment This command infers a consensus of 3 degrees of nucleotide identity with weighted conservation thresholds of 97 90 and 75 4 levels of presence conservation of nucleotides whose identity is not highly conserved 38 with thresholds 97 90 75 and 50 and tolerance for up to 10 of sequences having non canonical base pairs
83. emo breakpair varlen sto err B output of R2R with an error The line number with the problematic command is circled in blue and corresponds to the var_backbone_range command an error one or more base pairs is getting broken and you must choose how to resolve the problem Figure 3 5 gives an example of this problem and shows 4 ways to resolve it Pairs can also be broken when var_backbone_range commands refer to one side of a base pair In this case R2R will detect that this command was responsible for the problem Figure 3 6 If you have variable length stems you should use the var_stem or var_hairpin commands Key points e Pairs can be broken due to the consensus rules and Figure 3 5 shows 4 ways to resolve the problem e Pairs can also be broken due to variable length backbones e R2R error messages often refer to lines or columns in the input Stockholm format file The left most column is numbered zero The first line is numbered one 3 5 The place explicit command R2R encodes defaults for the layout of RNAs that work in many cases but often you will want to change this layout The place_explicit command allows you to position nucleotides relative to one another using explicit relative or absolute coordinates Figure 3 7 The syntax for the command is place_explicit label relative label place angle x relative y relative x absolute y absolute next angle illustrated with an example in Figure 3 7 The nucl
84. eotide at position label is positioned relative to the nucleotide at position relative label The coordinate system of z relative y relative is based on the direction of the nucleotide at relative label rotated by place angle degrees The x y coordinates x absolute y absolute are added to this position Both relative and absolute distances in the 20 Figure 3 7 The place_explicit command A 1 B STOCKHOLM 1 0 c 7 biologist AACCCUUCGGAAACGGAGUAA chemist AACCCUUGCUUCGGCGAGUAA demo pe physicist AUCCCAUUUCUUGAAGAGAAA c amp c b A H GC SS CONS KK gt gt K Ka eeen 0 0 A GC R2R_LABEL A Biss Cais DY _Icac gt c GF R2R place explicit aa 8010000 f 0 H GF R2R place explicit b b 45 1 90 H GF R2R place explicit c c 45 1 90 D mini ykkC O IILI o lo0000 700 O0000 lt 00 gt a 7 39 nt 1 228 nt 90th per centile 33 A Schematic of the positioning system in the place explicit command The diagram illustrates the effect of the command place_explicit G A 45 5 1 0 0 90 which defines how the nucleotide labeled G is positioned depending on the position of the nucleotide A The direction of the backbone at the A gray array pointing right is rotated 45 degrees to form the black arrow The position proceeds 5 drawing units along the black arrow and 1 unit orthogonally to the black arrow This is where the G is positioned The direction of the backbone of the G gray arrow
85. equences available for comparative analysis 78 6 2 3 How the file is modified by applying a predicate Obviously all sequences that match the predicate will be retained but all sequences that do not match the predicate will be removed Also obviously the GR and GS tags for discarded sequences will also be discarded Also all lines that begin GC R2R or GF R2R are removed Lines of the form GC SUBFAM_predicate name_other stuff text have the SUBFAM_predicate name_ part removed and same thing for GF SUBFAM_predicate name_other stuff text The special code GF SUBFAM_KEEPALL 1 means that all subsequent lines will be kept no matter what This is turned off with GF SUBFAM_KEEPALL O Also GF SUBFAM_predicate_name_KEEPALL keeps lines for the given predicate To show pseudoknot pairs as a special other diagram just define a predicate that matches everything but use the modifications to define a new GC R2R_LABEL line and probably other GF R2R commands You can also override the substitution string using GF SUBFAM_SUBST_STRING predicate name other string In this case the script will look for GC SUBFAM_other string instead of GC SUBFAM_ predicate name and same for GF This is useful for multiple predicates that all get drawn the same way Note the SUBFAM_SUBST_STRING command must precede the references the computer does not do multiple passes through the file 79 Chapter
86. er of every 10th nucleotide Default true indicateOneseqWobblesAndNonCanonicals If true then in oneseq mode see below R2R will indicate whether pairs are Watson Crick wobble i e G U or non canonical by using different bond symbols If false R2R will use the same symbol for all pairings specifically all pairings will be drawn as if they are Watson Crick Default true e pairShadingColors covariationColor compatibleColor noMutationsObservedColor change the colors used to shade pairs Colors are specified using standard R2R codes Good colors for slides are SetDrawingParam pairShadingColors rgb 156 199 153 rgb 152 199 222 rgb 235 138 126 thanks to Kirsten F Block for these values e solverMaxIters sets the maximum number of iterations of the solver i e CFSQP for layout of multistem junctions Sometimes a high number of iterations is necessary since the solver gradually creeps towards the solution However usually fewer iterations are needed and so the conservatively high default value is a waste of time Defualt 100000 Note you can compile R2R with a new default by adding DMAX_SOLVER_ITERS to the g command line e autoBreakPairs if this variable has the value true then R2R will automatically separate base pairs where one of them should be deleted and the other should not This is suitable for quick drawings where you don t want to think about how to resolve these issues If this variable has the val
87. er the GNU General Public License For details see http www gnu org licenses 1 5 Changes from previous versions e Version 1 0 3 August 2012 R2R 1 0 3 cannot exactly reproduce older drawings of RNA structures If you want to reproduce the old drawings you should use R2R version 1 0 1 The way that R2R handles degenerate nucleotides when calculating the consensus se quence has changed Degenerate nucleotides are now ignored in the calculations where previously they were treated as a uniform distribution of the possible nucleotides e g degenerate nucleotide R was treated as 50 A and 50 G As a result the current R2R does not exactly reproduce old drawings In some cases the layouts change when the differences lead to nucleotides falling above or below the 50 present level thus adding or removing nucleotide positions from the drawing Added ability to use alignment positions as labels See Section 5 7 5 Added dumpInfoFile drawing parameter which dumps information on layout to a tab delimited file Added section on interacting with R2R using scripts See Section 5 11 e Version 1 0 2 Added citation of R2R paper in BMC Bioinformatics to this manual Edited some error messages to make them more clear Check for conflicting use of bulge and place explicit Added make_pair command Added DNA shadeAlongBackboneWidth and alongBackboneMidpointGapLength draw ing parameters
88. er1 1 B Illustration of stem numbers and directions from the diagram of part A The enclosing stem is sO then the stems are number si s2 from 5 to 3 around the multistem junction The stem sO defines the angle zero as indicated As usual within R2R positive angles are in the clockwise direction and negative angles are counter clockwise Thus s1 is oriented in the direction 90 going to the left while s2 is in the direction O up the page If the enclosing stem stem s0 were rotated the angles of stems s1 and s2 would also rotate to be the same relative to the enclosing stem The angle of the enclosing stem stem s0 can be changed which has the effect of rotating the entire picture This is purely for convenience and is illustrated in Figure 3 11 C A variation on the drawing in part A The ai 0 parameters directs R2R to automatically decide on the intersection point with the circle for each stem This leads to a circle that better goes through each nucleotide D Drawing of the SAM IV riboswitch 16 using the solver In this case the bulged Y within stem s1 is very close to the A in the junction between stems sO and s1 The positions of these nucleotides could be adjusted in Adobe Illustrator of Inkscape Alternately the stem s1 could be rotated to avoid this problem as demonstrated in the next sub figure E Drawing of the SAM IV riboswitch with an alternate angle for stem s1 25 Figure 3 11 Automated layout of multiste
89. etters don t show up you might need to re create the SVG output with a different font name Please see the next section more technical information to learn more about this and Chapter 2 for information on how to change the font 11 3 1 2 1 More technical information R2R will generate valid PDF and SVG files as output In principle any program should be able to read them However I am exploiting the fact that the Helvetica font is built in to PDF Some versions of Inkscape will not render any text using PDFs built by R2R because they don t have the Helvetica font Therefore R2R can create SVG output which is Inkscape s native file format The SVG output uses the Bitstream Vera Sans font which is similar to Helvetica and to Inkscape s default font Meanwhile at least on Windows Illustrator will substitute Arial which is similar Note in principle you can substitute other fonts even by creating scripts to edit SVG output However R2R knows the heights and width of font symbols which is uses for its layout Thus for example if you use a narrower font the spacing might not be appropriate 3 1 3 Where the demo files come from Files within the demo directory were made to support this user manual Most are based on previously drawn RNAs RNA drawings derived from previous studies are organized into subdirectories The subdirectories and relevant citations are as follows e demo 22 Weinberg et al 2007 13 and Sudarsan
90. h alternate value for pairBondWidth to be used in skeleton mode Note that this is before scaling is applied Default 0 5pt skeleton backboneWidth alternate value for backboneWidth to be used in skeleton mode Note that this is before scaling is applied Default 1 5pt which is the same as the default for backboneWidth skeleton_outlinePseudoknots enable drawing an outline for callout style pseudo knots If this variable is set to true then commands ignore_ss_except_for_pairs outline will result in an outline even in skeleton mode If the variable is set to false then this command will not result in an outline in skeleton mode I think you usually don t want the outline Default false e Parameters relevant to circle_nuc or implicit circling from the GR CLEAVAGE line in oneseq mode 48 nucShrinkWithCircleNuc when the circle_nuc command is used nucleotides are shrunk to accommodate the circles They are scaled by a factor of nucShrinkWithCircleNuc Default 0 8 pairBondScaleWithCircleNuc scaling applied to base pair bonds to accommodate circled nucleotides Default 1 e Parameters relevant to oneseq mode drawing a single RNA molecule drawStandardCleavage Boolean true or false If true R2R will draw the standard cleavage diagram based on the GR CLEAVAGE line Default true defaultOneseqLabeling If true then in oneseq mode see below R2R will label the numb
91. has changed recently Put the file cfsqp c into the R2R directory NotByZasha cfsqp Then cd into this directory and run make 2 4 Make R2R 2 4 1 Optionally edit Makefile 2 4 1 1 Optionally enable CFSQP R2R is configured by default to disable CFSQP To enable CFSQP go within the src directory open the file Makefile in a text editor such as emacs or vi Remove the line that reads DISABLE_CFSQP 1 2 4 1 2 Optionally choose an SVG font If you intend to use Adobe Illustrator or CorelDRAW you can use the PDF Adobe Acrobat output of R2R In this case you do not need to set a font If you intend to use Inkscape you should use the SVG output of R2R since Inkscape is not reliable when importing PDFs generated by R2R By default R2R is set up to produce SVG output using the Bitstream Vera Sans font which is freely available http www gnome org fonts and might already be installed on your system If however Inkscape is not mapping your font correctly you can get R2R to generate SVG files that reference a different font Note it might be easiest to skip this step initially and change the font later if there are problems in Inkscape To change the font later modify the Makefile as described below then run make clean then run make To change this font open the file src Makefile in a text editor and change the definition of the FONT_CFLAGS variable Search for FONT CFLAGS and read the comment immediately above this lin
92. he SS_cons_into ss line replacing it with whatever is in SS_cons_replace ss and deleting SS_cons_replace ss primary is a special name that corresponds to SS_cons i e the primary secondary structure This is useful because Emacs RALEE gets upset with lines that end in whitespace Useful for drawing a pseudoknot pairing where the pseudoknot pairing should be the main pairing 65 5 8 4 7 merge_ss merge_ss replace ss into ss Same as subst_ss but merges the base pairs in instead of clobbering WARNING the code does not check if the base pairs are compatible i e if they re pseudoknots relative to each other This command is useful if you want to keep a terminator as a separate SS_cons which means that when you use the GC ACTUAL_MOTIF line you won t get only half of the terminator which would make the structure line invalid because it would have unmatched brackets However by using merge_ss you can put the terminator into the drawing with R2R 5 8 5 variable length regions 5 8 5 1 var_hairpin var_hairpin leftpair rightpair Replace the terminal loop contained within leftpair and rightpair with a variable length hairpin symbol leftpair and rightpair must be base paired with each other so I don t know why I require the user to specify both 5 8 5 2 var term loop var_term_loop leftpair rightpair message leftpair and rightpair must be the matching base pairs of the last base pair before the terminal loop
93. he base pair Any number from 0 to 1 can be given so ai 0 5 means start at the midpoint The circle intersection parameter is only relevant to multistem_junction_circular_solver the multistem_junction_bulgecircley commands try to have both nucleotides on the circle However other parameters are relevant to the bulgecircley variant commands You can also add try_harder to the multistem_junction_ circular_solver command to improve it s ability to find a good global optimum details in Sec tion 5 8 3 2 3 6 3 1 Solving multistem junctions can be slow On even modern machines the computer can take a significant amount of time to solve the problem Often the solver goes through a period of very gradually improving the solution and many iterations are required to reach an optimal solution and the difference in the ultimate solution is noticeable 28 Figure 3 14 An internal loop with a 90 degree turn A B E turn_stem_at_internal K 1 r E C Gy A G c right C G _ Ec C G i Es GC Cc 6 ue E 11 Cc G A 5 7 RCG G C C G o left A Ue Ye e 0 K A G C G Ve SAVANE A E ORCGACYRG a a YA A G GY ammm 0 C multistem_junction_circular_solver UEA a K s ai s1 90 ai EECA 52 GGYY Cc G G C 4 AU GCUGRY D multistem_junction E multistem_junction Yor SENERS EC bulgecircleynormals_solver _bulgecircleynormals cc K s ai s1 9 ai ELSE _solver G C Cc i i C 4 K s ai s1 45 ai ns Ce acucryc y GCUGRYC
94. he line is annotated with the range in the number of nucleotides contained in those columns which is au tomatically calculated For both commands the range is represented by two labels each of which refer to a column in the GC R2R_LABEL line For example the var_hairpin command used the labels lt and gt each of which refers to a specific column as shown Figure 3 2A Finally the drawing uses the tick_label command which labels a specific nucleotide position In this case I imagined that a particular G nucleotide has a special biochemical significance in this fictitious RNA structure Key points 15 Figure 3 2 demol ii Some annotation for the consensus diagram A B STOCKHOLM 1 0 var herein human ACACGCGAAA GCGCAA CAAACGUGCACGG chimp GAAUGUGAAAAACACCA CUCUUGAGGACCU demo1 ii bigfoot UUGAG UUCG CUCGUUUUCUCGAGUACAC bebe atsal GC SS_cons lt lt lt 0 ee O ee Un ial GC R2R_LABEL 0 0 Q 0 E PS we GF R2R var_hairpin 3 C i dash gt Ys rasce GF R2R var_backbone_range 1 2 character G C 0 0 GF R2R tick_label 0 magic G 5 ROO AC 2 3 nt label references space character var_backbone_range A Stockholm format alignment that includes a labeling line beginning GF R2R_LABEL and associated R2R commands beginning GF R2R Annotations in blue show how R2R commands are associated with a symbol in the labeling line Annotations also mark the special dash sym
95. he second messenger cyclic di GMP Science 321 411 413 2008 J X Wang E R Lee D R Morales J Lim and R R Breaker Riboswitches that sense S adenosylhomocysteine and activate genes involved in coenzyme recycling Mol Cell 29 691 702 2008 88 13 14 15 16 17 Z Weinberg J E Barrick Z Yao A Roth J N Kim J Gore J X Wang E R Lee K F Block N Sudarsan S Neph M Tompa W L Ruzzo and R R Breaker Identification of 22 candidate structured RNAs in bacteria using the CMfinder comparative genomics pipeline Nucleic Acids Res 35 14 4809 4819 2007 Z Weinberg and R R Breaker R2R software to speed the depiction of aesthetic consensus RNA secondary structures BMC Bioinformatics 21 3 2011 Z Weinberg J Perreault M M Meyer and R R Breaker Exceptional structured noncoding RNAs revealed by bacterial metagenome analysis Nature 462 656 659 2009 Z Weinberg E E Regulski M C Hammond J E Barrick Z Yao W L Ruzzo and R R Breaker The aptamer core of SAM IV riboswitches mimics the ligand binding site of SAM I riboswitches RNA 14 5 822 828 2008 Z Weinberg J X Wang J Bogue J Yang K Corbino R Moy and R R Breaker Com parative genomics reveals 104 candidate structured RNAs from bacteria archaea and their metagenomes Genome Biol 11 R31 2010 89
96. ide is also sometimes used for single stranded regions within a hairpin that should be positioned on a straight line instead of on a circle e InternalLoop internal loop or bulge At this stage also includes some outside regions that are between but not contained in stems These will be fixed to type Outside later e TerminalLoop 5 10 4 place explicit links The next section of the file shows how the drawing units are linked together by place_explicit commands or by implicit rules For example here is one item of output ssContext 147 172 inf raw 0 23 109 109 Outside link posFrom 22 147 172 inf raw 0 23 109 109 pasto 23 173 177 278 282 raw 23 28 82 87 default rule 45 1 0 90 The ssContext line introduces a drawing unit whose links will be listed The raw specification and the meaning of Outside were explained above The text 147 172 inf defines the drawing unit in terms of text columns within your input alignment file As always the left most text column is numbered zero to follow emacs The equals sign says that the right part of the drawing unit is empty The range 147 172 is a doubly closed interval that includes text columns 147 and 172 and the columns between them The link introduces a place_explicit style link showing how the drawing units can be posi tioned relative to one another posFrom 22 specifies that raw position 22 will be the nucleotide position within the firs
97. ight line like the layout_straight command note you re responsible for making sure the length is good bpez is an independent command that really only makes sense with bsi and it s a place_explicit command that positions the bulge itself you ll usually have to use it with bs since otherwise the bulge will go 90 degrees A short form for this is bss1 which both says that the bulge should go straight and that it should be positioned straight from the base pair it s the same as bsi combined with bpez 0 1 0 0 0 0 blinearstretchi says make the bulge straight and stretch it between the nucs it has to go btrianglei corner says make the region into a triangle really two linear segments with the corner at the position identified by the label corner which must be within the bulge btrianglefi is the flipped version i e the corner point goes on the opposite side of the direct line e bspecifyendpointsi relDir beforeBulgePos X beforeBulgePos Y afterBulgePos X afterBulge Pos Y 58 Specify the positions of the endpoints of the bulge the bulged nucleotides will be positioned between them The endpoint positions are relative to the position of the left nucleotide of the enclosing base pair in coordinates oriented at relDir This is mostly useful for automated layout commands generated by R2R s solver e bdrawcirc Says that the computer should draw the full circle related to bulge 7 which is useful for debugging or for doing to
98. ion it is often necessary to play with several formulations involving different stem angles and solvers One strategy to finding good directions for the stems is to use the default circular positioning of the multistem junction to find directions that fit the circle perfectly then consider nearby angles that are multiples of 90 degrees or have other pleasing characteristics See Figure 3 11 for an example As noted above the starting point for the circle intersection point of each base pair can sometimes have an effect on the solution Variable length backbones introduced using the var_backbone_range command can help to improve aesthetics of drawings as the solver is permitted to change the length of the lines to improve the circularity of the layout This additional degree of freedom can be helpful Of course some multistem junctions do not have variable length regions so this tactic cannot always be used Additional examples illustrating aspect of the solver commands are shown in Figures 3 12 and 3 13 Note the starting point for the intersection can be supplied for automatic inference and this sometimes assists the solution as the problem that the computer has to solve is highly non linear and the computer can easily get stuck in local minima See Section 5 8 3 2 for details on various initial values that can be set For example ai 0 means start with the intersection at the left nucleotide and ai 1 means start at the right nucleotide of t
99. le of an internal loop There are three ways to draw multi stem junctions with R2R e The junction is laid out perfectly on a circle and stems are allowed to go in arbitrary directions This is the default It usually looks acceptable except when nucleotides clash although I think it s typically not the most aesthetic solution e Manual layout using place_explicit commands e Automated solutions that attempt to position the junction on a circle as well as possible but subject to constraints on the directions of the stems This allows for orthogonal stems a design feature that I think tends to result in more attractive layouts 3 6 1 Simple circular layout The default circular layout for multistem junctions is demonstrated in Figure 3 8 This is the layout that R2R uses if you do not give R2R any instructions on how to draw the multistem junction 3 6 2 Manual layout of multistem junctions Manual layout of multistem junctions can be accomplished by positioning the stems using place_ explicit commands and directing the layout of the single stranded regions using the bulge com mand Since the use of these commands would result in many labels to identify the various positions R2R has the multistem_junction_bulgey command for which only one label needs to be defined 22 Figure 3 8 Simple circular layout of multistem junctions multistem_junction_circular M allstems any angle flipstem 1 m stem 1 REY R UsRA U bad nt
100. letters so its width is implicitly set by the nucleotide font size Because R2R is not programmed to handle all cases there will be abrupt changes in the direction of the shading line when joining nucleotides in some directions These abrupt changes are usually not very noticeable Note you can use the shading produced to highlight a set of nucleotides in other ways The shade_along backbone command creates a complex path series of lines and arcs that can be manipulated in drawing programs Also for example in Adobe Illustrator select the line and use the Outline stroke command under the Object path menu in Illustrator CS to convert the line to an outline shape which you can then manipulate 5 8 6 10 Outlining a stretch of nucleotides around both ends outline_along_backbone label label2 color This command has the same syntax and idea as shade_along backbone except that it creates an outline around the nucleotides There are limitations of this command First the method is somewhat of a hack it simply uses the code for shade_along_backbone to draw a larger shade in the outline color then uses the code again to draw a smaller white shade What s left is the outline and there s no need to calculate the actual outline The limitation is that you can t draw the outline on top of anything because the white out shading would obscure whatever s below Therefore the outline is drawn below any covariation shading This of
101. lge is placed between the position immediately before the nucleotide at label and the position immediately after the drawing unit containing label In earlier versions of R2R the user could set the before amp after points explicitly but I found this was never needed the added flexibility just allowed the user to draw things in the reverse direction and make errors If N single stranded nucleotides are laid out as a bulge from nucleotide X to nucleotide Y then the computer calculates a circle such that e X and Y are points on the circle e The distance between all consecutive nucleotides along the region from X to Y including the single stranded region label is the constant internucleotide length bulge_flip label Same as bulge which was just discussed above but draw the circle on the other side of the line from nucleotide X to Y 5 8 2 3 Layout single stranded loops along a straight line instead of circle layout_straight label This command forces drawing units that would normally be circular bulges internal loops and terminal loops to be straight Usually the command is not necessary because place_explicit commands that intersect circular drawing units will cause them to be straight 5 8 2 4 turn_ss turn_ss label turnAngle Turns a single stranded region by the given turnAngle The command is equivalent to the following place explicit command place_explicit label label turnAngle 2 1 0 0 0 turnAngle 5 8 2 5 tu
102. list of parameters that can be modified with SetDrawingParam is provided in Section 5 5 2 e Lines beginning GC R2R_XLABEL something can be used to introduce an extra dimension of labels with the name something See Section 5 7 for details 3 4 Common error One or more pairs is getting broken by one side getting deleted This sections explains how to handle a common error when using R2R R2R defines which nucleotide positions are shown in a consensus based on how frequently they are present in sequences Sometimes one nucleotide position involved in a base pair is removed from the consensus while the other is retained This can happen with stems that are loosely conserved When this arises R2R will report 18 Figure 3 5 demo breakpair common error B ERROR there was a problem drawing motif demo ALEA AT breakpair One or more pairs is getting broken by one VACCCCUAU side getting deleted presumably because it s not eee ss conserved while the other one stays Because of mire AAGCAACCCCUU CGG generic process probably RemoveGaps and not the ACES RE st result of any specific command Consider using GF R2R keep allpairs Note another explanation is that you ve used GF R2R var_backbone_range on columns that includes a pair text columns in alignment 16 25 left keep right chuck left context AARC PP 1 0 D STOCKHOLM 1 0 lemur AAACUACCCCUAUUGG lemur AAACUACCCCUAUUGG blueberry AAACAUCCCCAU UGG
103. ly non linear problems However it is difficult to find such a starting point in an automated fashion The current starting value for the center of the circle seems to work well With _solver even slight changes in the initial fraction of stem intersection e g the number in the ai option for stem layout strategy can be the difference between success and failure I find the bulgecircley commands to be more robust but it s sometimes necessary to try multiple versions of the solver commands and fiddle with parameters The commands differ as to their strategy for solution multistem_junction_circular uses a highly limited version of the problem and solves using a simple binary search over the problem defined in one variable the height of the circle along which the multistem junction is positioned With _solver a more general version of the problem can be posed in multiple variables and CFSQP is used solver will generally take much longer to compute 99 The problem the commands solve is also different multistem_junction_circular and multistem_ junction_circular_solver attempt to solve essentially the same problem all nucleotides in single stranded junctions are forced to be on a circle and the computer attempts to fit the base paired nucleotides to be close to the circle The best solution is the one with the smoothest transition between stems and junctions Mathematically the computer is minimizing the sum of squared devi ations
104. m junctions unusual stem directions AA AAA B G A R Y R gt OU CA A y A TA c Vee C A yy RA AU a yl ARYUCA vee Y uma PYRAJU O AUB G Mn Ve oe Ur y0 7G O O eo WZ O G se Y AS e e c o uce Ge G C 6 U ec eo U A AA ou 0 0 AA 5 multistem_junction_bulgecircley_ solver j se 15 ai s1 135 ai s2 45 ai s3 45 ai s4 105 ai C zero degrees AA G A Y R e nucleotide referenced A0 0 y by align Y AL00 Cc 11 yoA it Ayoo aang S YRA U G UME AU Uo o nucleotide referenced O by align X usc ce O o 90 AA multistem_junction_bulgecircley_solver j se 15 ai s1 135 ai s2 45 ai s3 45 ai s4 105 ai align_nuc_centers_angle 60 align X align Y align_angle 30 0 4 This figure illustrates the strategy of using the default circular layout to suggest aesthetic stem direc tions In this case I was not able to find directions that are multiples of 90 or even 45 that worked well This example is based on a multistem junction in the manA motif 17 see demo 104 manA sto for the full motif A Default circular layout of this motif This drawing is derived from the file demo demo multistem manA sto B Layout using solver with stem directions that roughly match the default orientations in part A Stem sO is rotated 30 degrees clockwise in this figure The directions of stems sO and s4 are symmetric about the vertical axis which results in an aesthetic design Note R2R does not draw the 5 marking since the 5 nucleotide s
105. mol r2r_meta output demol svg Key points e R2R inputs are Stockholm format files The Stockholm format is explained more at http en wikipedia org wiki Stockholm format e Consensus secondary structure is expressed using brackets e g lt and gt in the GC SS_ cons line e g Figure 3 1 e R2R will infer the consensus sequence and classify covariation e Creation of the output requires two UNIX commands 3 3 2 Some typical customizations for consensus diagrams In this section we will use some R2R commands to improve the consensus diagram drawn in the previous section This will entail adding an GC R2R_LABEL line which will allow us to label and refer to specific columns in the alignment The demol drawing in Figure 3 1B has 5 and 3 flanking regions that are not conserved according to R2R s definition of conservation Although it is often desirable to keep such poorly conserved flanking regions in curated alignments they add little to a diagram By adding dashes to the R2R_LABEL line we cause R2R to remove those columns in the output see Figure 3 2 The first hairpin of the demol motif varies in length and its terminal loop is poorly conserved The R2R var_hairpin command allows us to replace part of the hairpin with an abstract rep resentation of a variable length hairpin Similarly a part of the junction between the hairpins is poorly conserved and the R2R var_backbone_range command replaces this with a line T
106. n then R2R will never declare covarying compatible or non mutating pairs In other words it will always declare code which corresponds to base pairs that are not shaded in consensus diagrams A concrete example of these parameters and their meaning was given in Section 4 1 Note if zdentity levels is greater than 3 or present levels is greater than 4 R2R is not designed to be able to draw the consensus diagram although it would be relatively straightforward to extend it to interpret additional levels I have assumed this is not likely to be an important feature 4 5 Output of R2R for the consensus You probably won t have to understand R2R s consensus output except to understand that R2R removes gap columns when it draws the consensus though you can override this with the keep command and that it annotates these gap columns in the line beginning GC cons R2R s consensus output consists of a series of GC lines that annotate per column data These lines are e GC cons Consensus sequence Gaps are represented as dashes Nucleotide symbols A C G U represent conserved nucleotides Two degenerate IUPAC symbols are used R means the position is conserved as either A or G and Y represents either C or U A lower case n 40 is used for columns that are typically or always present but do not conserve the nucleotide identity These positions are drawn as circles by R2R An annotated example is given in Figure 3 1D
107. n the drawing whether each element is flipped or not Another strategy is to simply remove the f marking from place_explicit commands that you might think should be flipped and see if that helps Finally you can use the command SetDrawingCommand showPlaceExplicit true to show you how your place_explicit commands were used in the positioning and where default rules were used which you might want to override with additional place explicit commands The showPlaceExplicit option is demonstrated in Figure 3 17 Examples of R2R based RNA drawings with in line pseudoknots are as follows Most of these use flipped stems 34 Figure 3 17 A pseudoknotted RNA drawn in inline style A AAy_pYUUAURYCAAGUUUA R Y R A U G C C G G C e 5 PTYARCAU OGY RAA B STOCKHOLM 1 0 rope AGGCAUUUGAACCAUAUUGUGCGCCUAACAUC GCCAAAGCACAA speed GGGUAUUUGAACUGUAUUAUGCACCCAGCAUAAUGUGGAACCAUAA topology AGGUAUUUGAACCGUAUUGUGCACCUAGCAUGA GUUAAAGCACAA GC SS_cons EEE spas di E E 6C SS CONS L gaa a Ed gt gt gt gt GC R2R_LABEL Satin A aves EEEE Taea Drese A 3 H GF R2R place explicit t t 10000f H GF R2R place _ explicit 3 3 45 1 90 f GF R2R var_backbone_range 1 2 GF R2R bulge a GF R2R bulge flip b C demo pknot inline D E RUA RUA J o YU AA o vu A A A R Y E R Y 1 1f A U A A U A G G C G G U C AAR U AAR Uy A MER e Uy Aver gt def G C ES G C 5 G C G C ES 6 m U A 5 A 3 nt a 3 nt
108. nds have been resolved into place_explicit commands and is implemented in the file PositionBackbone cpp 8 2 2 Inferring a path for the backbone In order to draw a skeleton style schematic drawing or for shading consecutive nucleotides e g using the shade_along_backbone command it is necessary to determine a path corresponding to the RNA s backbone In R2R this path is composed of straight lines or arcs The path largely follows the lines and arcs used to position nucleotides Except when certain commands are used all nucleotide positions in R2R are laid out either on a straight line or along a circle where each nucleotide is centered on the endpoint of the line arc For each position the computer stores the line or circle defined at that point However it is not sufficient to use these path segments to connect the diagram as they are only defined at the nucleotide itself and not for the connection to the next nucleotide Moreover even when two nucleotides are connected the connection might not be aesthetic Some nucleotide positions are not really points For example the var_backbone_range command results in a line or arc even though it is represented as a nucleotide position Therefore the first step is to expand nucleotides into separate points We call these anchor points by analogy with the drawing of complex curves in programs such as Adobe Illustrator Thus we are given a set of anchor points a1 d2 Each anchor poin
109. neseq for a predicate defining a modular structure GF Makefile oneseqpred hit predname Directs the script to draw a subfamily defined by the predicate predname in oneseq mode for the sequence named hit Optionally define directives can appear after predname The predicate predname must be defined earlier in the file 7 2 2 6 Weighting sequences using your own algorithms with MetamakeDemos To use manual weighting for a given Stockholm file add a line like GF Makefile SeqWeighting command where command is a UNIX command to create a new Stockholm file with the weights Within command the special text INPUTSTO will be changed to the input Stockholm file and similarly OUTPUTSTO will be changed to the output Stockholm file The result of the command should add the GF USE_THIS_WEIGHT_MAP tag to the output Stockholm file as described in Section 4 3 If your Stockholm file already has a valid GF USE_THIS_WEIGHT_MAP tag in it you can use the following line GF Makefile_SeqWeighting cp INPUTSTO OUTPUTSTO which just copies the input file that already has the tag to the output file 83 Chapter 8 R2R source code 8 1 Summary of C and Perl source code files AdobeGraphics cpp Implements generic aspects of interface to draw in graphical files such as PDF format files AdobeGraphics h Interface to draw in graphical files AdobeGraphicsLayout cpp Implements generic utility classes to assist in drawing and lay out
110. nment file D The Stockholm format alignment file generated by R2R with the GSC weighted consensus flag that has information on the consensus of the demol motif Some minor changes were made to improve readability Note that it is not necessary to understand the information in this file because the file is simply used by R2R to produce its final output The way in which R2R calculates consensus diagrams and the meanings of the colors and symbols is explained in the research paper as well as in Chapter 4 E The contents of the file demo demo r2r_meta In this case the file simply lists the name of input file used to draw the consensus 14 commandprompt src r2r GSC weighted consensus nput file sto output file sto 3 0 97 0 9 0 75 4 0 97 0 9 0 75 0 5 0 1 This command is explained in Chapter 4 and the parameters are fully explained in Section 4 4 Also while it is necessary to generate consensus information so that R2R can draw it users can write their own routine to generate the consensus data instead of using the GSC weighted consensus command this alternative is explained in Section 4 6 Given the consensus data generated by the previous command we run R2R to create a PDF file The file demo1 r2r_meta simply gives the path of intermediate demo1 cons sto which is what R2R will use to draw commandprompt src r2r demol r2r_meta output demo1 pdf Alternately to create SVG format output commandprompt src r2r de
111. ns are only detected in drawing 5 10 2 consensus lines and raw coordinates Next R2R will draw the consensus sequence line after columns are removed because of gaps explicit deletion or variable length functions The codes for the consensus line e g nnnRGGACGACC are explained in the previous chapter for the GC cons line The consensus line defines the raw coordinates which are numbered from 5 to 3 starting at zero Next all SS_cons lines are output again after removal of columns 5 10 3 Preliminary secondary structure drawing units Next R2R will dump its preliminary secondary structure units called ssContextList This is immediately after parsing and does not take into account any splitting of drawing units that will be done later for place_explicit commands For each unit it will print something like this 0 8 12 20 T F T Pair 72 In this example the 0 8 12 20 defines the nucleotide positions of the drawing unit in raw coordinates 0 8 is the left 5 side of the unit and is a half open interval In other words it is the set 0 1 2 3 4 5 6 7 thus including the first number 0 but not including the second 8 Similarly 12 20 defines the right side of the unit and is also a half open interval The T F T defines flags that are not important The Pair defines the type of drawing unit The following are types e Pair stem e Outside single stranded region outside a pair Outs
112. nto a new sto file using a Perl script that is part of R2R and 3 draw the new file Predicates can be defined using either regular expressions or using Boolean operations using a subset of the Perl programming language Regular expressions are a mathematical formalism for describing sequence patterns 1 Some examples are given in Figure 3 15 To implement these regular expressions I used Perl and conse quently R2R uses the Perl syntax to specify regular expressions To learn about all the capabilities of regular expressions in Perl one good Web site is http www perl com doc manual html pod perlre html Regular expression predicates operate on subsequences in columns defined by lines of the form GC SUBFAM_LABEL_ For example GC SUBFAM_LABEL_TERM would define columns that can now be referred to as TERM Here are some examples of regular expressions that illustrate features relevant to RNA For four columns with no gaps G A Z AG A defines the GNRA tetraloop The text in square brackets and define a set of symbols A Z is the set of upper case letters while AG defines the symbols A or G A period is a special symbol that stands for any character so in order to look for a gap character you should enter A few other special symbols are often relevant The caret symbol shift 6 on a standard US keyboard stands for the beginning of a subsequence while the dollar sign represents the end of
113. ollowing command line options which are not normally used 80 srcFileList filename directs the script not to search for sto files Instead filename gives the list of input sto files Each line in filename specifies an input file and is tab delimited The first field is an actual source sto file For convenience of Windows CygWin users any backslashes in the file name are converted to frontslashes The second field in the line is a name for the file The Makefile will copy the source sto file into a new file called intermediate name sto where name is the second field makefile filename creates the new Makefile with the name filename The default is makefile Makefile r2r executablefilename specifies the command to run for r2r Default is r2r src r2r subfam filename references the script SelectSubFamilyFromStockholm pl Default is subfam src SelectSubFamilyFromStockholm pl noDumpFile do not get r2r to dump solutions of multistem junctions noall do not create an all target in the new Makefile This is useful if you want to include the Makefile generated by MetamakeDemos p1 into some larger Makefile doStoTex experimental feature doStoOutput experimental feature pubsto perl file experimental feature cleansto perl file experimental feature 7 1 2 Makefile targets The script generates a Makefile with the following targets all equivalent to all pdf and all svg Not recommend
114. on number Add G residues to the 5 end of the sequence in the alignment The number of Gs to add is given by number Each G has an asterisk added to it as if with the tick_label command All of these added Gs gets the special label g_with_transcription so you can refer to them by this label Non genomic G nucleotides are often added to the 5 end of RNAs because it increases the yield of in vitro transcription using T7 RNA polymerase 5 8 7 4 Keeping gap columns in the drawing keep label1 label Forces alignment positions corresponding to any of the listed labels to be kept in the alignment even if they would normally be considered gaps and removed 5 8 7 5 Breaking pairs depair label other labels Converts the columns matching any of the given label s to single stranded It will be as if these nucleotides were never annotated as base paired For example this command removes all base pairs in the GC SS_cons_1 line depair _1 lt _1 gt 5 8 7 6 Deleting columns using an explicit command delcol label1 label2 Deletes the columns specified One more more labels may be specified The effect of this command is equivalent to putting a dash character in the R2R_LABEL line except that the delcol command can be applied conditionally using ifdef or ifdefeq commands 5 9 Troubleshooting NOTE R2R will parse the RNA structure into units and commands only apply to the first nu cleotide in a unit See Section 5 3
115. on relativeLabel angle placeAngle where relativeLabel angle is the angle of the backbone at the nucleotide specified by relativeLabel We then travel relative Vector X units in this direction Then we rotate the unit direction vector by 90 degrees and travel relative Vector Y units The angle of label is computed by taking the angle of relativeLabel and adding finalAngle It is possible to use place_explicit to place the right nucleotide of a base pair In this case remember that the backbone for nucleotides on the right side 3 side of base pairs is considered to go in the opposite direction as 180 degrees from the nucleotides on the left 5 side place explicit can also be used to place the last nucleotide in a structural unit like the last nucleotide of a contiguous stem If the optional f flag is given at the end of the line the left rightness of label is flipped relative to that of relativeLabel Note that place explicit commands can be evaluated in both directions in a mathematically equivalent way In other words if you position label A based on B there is an equivalent place explicit command to position B based on A R2R will pick whichever layout order is convenient 5 8 3 Layout of multi stem junctions 5 8 3 1 Manual layout multistem_junction_bulgey label see below This is a convenience function that allows you to render a multi stem junction by placing each stem in the junction relative to the previous stem
116. on with examples on how to create RNA drawings with R2R Example input files containing R2R commands to draw various RNA structures are available in the demo subdirectory Credit the Stockholm format input files in the demo directory and its subdirectories are similar or identical to supplementary data published in previous reports 13 16 9 12 17 15 6 3 1 About the demo files 3 1 1 Demo output is already provided I have created the raw output of R2R when run on the demo files and this output is provided within the output pdf PDF format and output svg SVG format directories If you just want to see the output I recommend opening the PDF files using Adobe Reader http get adobe com reader as this should work on any platform Note if circles look very unsmooth when viewed in Adobe Reader please see Section 2 5 If you want to try editing the output files please see Section 3 1 2 next to find out which files to use If you want to create the output files yourself see Section 3 11 3 1 2 Drawing programs Adobe Illustrator or Inkscape Simple conclusion e If you re using Adobe Illustrator or CorelDRAW use PDF output The example commands below use PDF e If you re using Inkscape use SVG output In the commands below wherever it says pdf instead write svg R2R decides the output format based on the file extension of its output file either pdf or svg If you have problem with the fonts in Inkscape or if the l
117. ons To direct R2R to output this information set the dumpInfoFile drawing parameter see sec tion 5 5 2 to the desired output file path The format of the file is as follows The file is tab delimited and any whitespace below is actually a tab character The file contains multiple sections Each section begins with a header line giving a name of the section All headers except for the first drawingName specify the number of lines contained in subsequent lines of the section So to read this file read a line to determine the section and number of lines of data then read that number of lines then read the next header line etc An R2R output file PDF or SVG can contain multiple drawings in it For example you might have a consensus diagram a consensus of one or more pseudoknots and a single sequence oneseq mode All of these drawings will appear in the same dump file The order of sections in the file is always the same except that the sections will be repeated for each drawing The sections and their order are as follows e Header line drawingName name The name of the drawing whose data will come next The drawing name is given in at the top of each drawing in R2R s PDF or SVG output There are no additional data lines in this section just the header line e Header line posToAlignCol num data lines R2R removes alignment columns from the drawing when they are gappy or as a result of some commands e g var_backbone_rang
118. ow cleavage decreasing red cleavage increasing green 44 5 5 r2r_meta file You specify what files R2R should process in a file with extension r2r meta The file is tab delimited If the first tab delimited field is empty the line is ignored i e this is a comment If a line has only one field that field is the path to a Stockholm format alignment file This file is processed to draw a consensus diagram If the line s first field is SetDrawingParam then you can change the default settings of certain drawing parameters see below You can also specify that a single RNA molecule should be drawn rather than a consensus as described below 5 5 1 Defines You can define symbols that control which R2R commands are processed as described in Sec tion 5 8 1 You can add define name value to any line that specifies a sto files in the r2r_meta file These defines affect the drawing of that sto file For example see the file demo SAM IV SAM IV r2r_meta This file directs the drawing of SAM IV riboswitches in various ways and was used for the various examples in the tutorial that used SAM IV Defines are only applicable to the line in which they appear 5 5 2 SetDrawingParam You can include lines like the following in the r2r_meta file SetDrawingParam name value to change internal drawing parameters in R2R This affects all subsequent lines of the r2r_meta file Valid name codes are e Miscellaneous font
119. pairBondWidth 0 5pt R2R if_skeleton shade_along_backbone s rgb 255 0 0 N A The r2r_meta file name demo demo1 iii r2r_meta which directs R2R to draw a regular consensus diagram first line a single sequence oneseq from the bigfoot sequence and a skeleton style schematic drawing of the consensus Note that the components in each line are separated by a single tab character B Stockholm format alignment This text comes from the file demo demo1 iii sto Some irrelevant markup has been removed The two lines beginning GF R2R_oneseq are only processed within oneseq mode i e only for the bigfoot sequence in this example They refer to column labels with alternate names bf and bf1 that are introduced using GC R2R_XLABEL_bf and GC R2R_XLABEL_bf1 In the oneseq case dashes in the R2R LABEL line are ignored but dashes in the GR bigfoot DEL_COLS lines cause nucleotides to be eliminated from the drawing The last line which contains if_skeleton is only interpreted in skeleton mode and it overrides a default drawing parameter to set the basepair bond width to 0 5 points C Raw output of R2R rearranged to fit the figure space efficiently From top to bottom is a standard consensus diagram a single RNA molecule and a skeleton schematic 17 Figure 3 4 Other kinds of drawings illustrated with SAM IV riboswitches A SAM IV NC_003888 3 2308784 2308334 B SAM IV skeleton with bp SoFD0008 G Og Ge G G U 4
120. probing experiments Examples of cleavage diagrams drawn using R2R are those of the SAM IV riboswitch 16 Fig 2 A and HEARO RNA 15 Supplementary Fig 8 b In oneseq mode see r2r meta file above two special GR tags are processed e DEL COLS columns with a dash will be deleted These are used to specify the RNA subsequence that was actually probed Note that the R2R_LABEL line does not specify deletion in oneseq mode e CLEAVAGE nucleotides within a column are colored according to the following code constitutive cleavage yellow decreased cleavage with ligand red increased cleavage with ligand green no data gray Anything else no cleavage not shaded 50 note you can also circle nucleotides with your own colors using the circle_nuc com mand e R2R_LABEL R2R_XLABEL see labels below If present these GR tags override anything in the global GC tags Note I find it easier to just use global GC R2R_XLABEL lines and reference them in sequence specific drawing commands 5 5 6 1 Multiple drawings of the same RNA molecule In oneseg mode you might have done multiple experiments with the same sequence and want to show distinct data with each experiment This situation would cause a problem because the hitId that is used with GR tags would have to be the same To avoid this problem you can add construct where construct is any string that specifies a given RNA
121. ps along a straight line instead of circle 56 eat MILI apra AA A DS ES 56 few turisten at Merial e ida se e 56 820 Pa Expl ge kk wk Rhea aa 57 Ded Layout OF multi stem JUNCUS soo e oe a AA a 57 Bee Manval layout o uoc ni s a ma e a DE RA e 57 5 8 3 2 Multi stem junctions automatic circular layout 59 5 8 4 Changing layout of secondary structure 64 DRAI dpat o oes rean oe See Yee eee eee ee ee a 64 5 842 Wea o e sac ee aden raras A ee Me 64 5 8 4 3 Internal loop to bulges 02000000008 65 5 8 4 4 Ignore pseudoknots entirely eee 65 5 8 4 5 Ignoring pseudoknots for the purposes of layout 65 ee SA at eG Atak fhe ye ee Be ee ee cs AA 65 EA cd ok ey eee we ee we eee Oe be a es 66 28 5 variable length regions lt lt 0 2200 ea a ee eR EEE HO 66 Gaal AECE y eee EA 66 Oe WHEE esc RIERA Aa 66 5 8 5 3 Variable length backbone 00000 ee 66 AA RIA 67 DEO ARMORROA so ses So ee nia ta a AE AR A O A 67 5 8 6 1 Tick labels like the nucleotide numbering in cleavage diagrams 67 woes OPA os ke kh knee have Ge PaGb Babe eee Pee MS 68 a803 Psendo hold tie Licda EWR EEE REE EEE 68 5 8 6 4 Changing nucleotide colors a eee 68 D805 Circling nucleotides lt s o ee AAA 68 58 6 6 Boxing nucleotides lt ucooscr orina rra kiris 68 5 8 6 7 Outlining inlining nucleotides lt 68 5 8 6 8 Boxing a set of nucleotides
122. r labels are intentionally small so that they clearly indicate the relevant nucleotides in an automated drawing Most PDF and SVG viewers allow zooming in to expand the scale of these features 35 e SAH 12 13 see demo SAH SAH sto e PreQ II 13 see demo 22 preQ1 I1I sto e The pfl motif 17 see demo 104 pf1 sto The ykkC III motif 17 see demo 104 ykkC III sto e The Downstream peptide motif 17 see demo 104 Downstream peptide sto 3 9 Preparing presentations using projectors This section concerns the default colors for shading base pairs which indicate covarying mutations These colors were selected with journal papers in mind The colors are clear in print or on a computer monitor yet are not too distracting However these colors are too light for many projectors Therefore if you re doing a talk and want the audience to be able to see the shading I recommend darker colors To modify the colors in an R2R drawing you can add the following R2R command SetDrawingParam pairShadingColors rgb 156 199 153 rgb 152 199 222 rgb 235 138 126 thanks to Kirsten F Block for these RGB color values Or you can change the colors of an existing drawing program in a drawing program Adobe Illustrator allows you to select all object that have the same fill color as a selected object So you can select one covarying pale green base pair and then get all the others to change this color 3 10 How to draw GOLLD RNA from R
123. rithm per file annotation e GC col entropy Experimental feature the entropy of each column written vertically e GF DID_FRAGMENTARY FALSE Use of FRAGMENTARY feature is recorded per file annotation 4 6 Generating your own alignment consensus bypassing R2R When drawing a consensus R2R depends on data added by the GSC weighted consensus com mand which is added to the original information in the Stockholm format alignment file An alter nate program to calculate the consensus data need only output these extra data fields The required fields are GC cons GC conss GC cov_SS_cons and for any extra stem GC SS_cons1 there must be GC cov_SS_consz The form of these fields is described in the previous section Examples can be generated from the demo files by looking at the output of the GSC weighted consensus command in the demo intermediate directory the file names will end with cons sto R2R is however committed to up to 9 conservation levels each corresponding to a different color For a larger set of colors the color of nucleotides can be overridden later using the nuc_color command see Section 5 8 6 4 There is no command to specify an arbitrary color for shading of base pairs 41 Chapter 5 Reference drawing 5 1 Running R2R R2R processes Stockholm format sto files with special markup The files must have information on the consensus which is also made by running R2R see Chapter 4 There
124. rn_stem_at_internal turn_stem_at_internal label dir optimize for Uses an internal loop to turn the stem 90 degrees label must identify either the 5 most nu cleotide in the 5 part of the internal loop or the 3 most nucleotide in the 3 part of the internal loop dir is either 1 or 1 and this value is multiplied by 90 and added to the current angle of the stem optimize for is either L or R which means optimize the sizes of the circles for the bulges based on either the left or the right sides of the internal loop Note if you have problems with this you can also try the multistem junction commands If it doesn t work try this split the internal loop into bulges with internal_loop_to_bulges make the bulges as bulges along a circle with bulge label and finally use place_explicit to manually set where the stem goes 56 5 8 2 6 place explicit place explicit label relativeLabel placeAngle relative Vector X relative Vector Y abso lute Vector X absolute Vector Y finalAngle f Positions a nucleotide relative to another label specifies the region that will be positioned Positioning is done relative to relativeLabel The position of relativeLabel is taken as an origin The following two positions are added to this number where all coordinates X and Y are specified in units of the distance between consecutive nucleotides e The vector absolute Vector is added directly e We form the unit direction vector in the directi
125. rrespond to 0 42 0 525 inches e The positive X direction is to the right the positive Y direction is down 74 e When nucleotides X and Y are base paired the position of Y depends on the orientation angle of nucleotide X and the drawing parameter defining the distance between paired nucleotides see pairLinkDist in section 5 5 2 Only one of the paired nucleotides should be positioned An exception is for pairs that are ignored e g by the ignore_ss command these nucleotides are not treated as being base paired by R2R for the purposes of determining the layout Use bulge or bulge flip see section 5 8 2 2 to position nucleotides on a circle These com mands will correctly set the orientation of each nucleotide in the bulge and will consider the backbone to lie on an arc It is also possible to set the position of each nucleotide individually using place explicit commands but R2R will not realize that the nucleotides are on a circle This will affect the behavior of some commands e g shade_along_backbone and var_backbone range Currently no command is implemented that would allow the input to specify a circle explicitly 5 11 2 Accessing layout information from R2R using scripts R2R can dump information on the data it will draw including which alignment positions were not drawn because they re gappy or because of a user command and the coordinates of the nucleotides The format is a tab delimited file with multiple secti
126. s oN Y AC G 5 C U C G G 0 0 5 G YY o 0 Es C G USA U A G A G A A 2nt A a A A G o e A e yc cu multistem_junction_circular_solver multistem_junction_bulgecircleynormals_solver j s m s1 45 m s2 45 m j s m s1 45 m s2 45 m s3 135 m flipstem 2 s3 135 m flipstem 2 This figure illustrates an additional example of a multistem junction drawing based on a multistem junction found in IMES 1 RNAs 15 The drawing is based on the file demo demo multistem IMES 1 sto A Layout of the given command B Layout a different solver command as shown Figure 3 13 Automated layout of multistem junctions starting point is important AA AA A UA B UA G C G C G C G C Ex U A GA U A Y yy at Y TT ut U a U Get T p o R Fi 4nt A G 5nt 4nt A G7 5nt G C G C A U A U ae STA multistem_junction_circular_solver multistem_junction_circular_solver J s O ai s1 90 ai s2 ai J s ai s1 90 ai 0 s2 ai 1 align_stem_horiz 0 2 align_stem_horiz 0 2 This drawing is based on a multistem junction in HEARO RNA 15 that is different from the earlier multistem junction shown The drawing is based on the file demo demo multistem HEARO2 sto A Layout of the given command with ai parameters using the default midpoint as the initial point for the optimization search B Layout of the given command with ai 0 and ai 1 specifying alternate starting points The resulting layout is slightly different and I think a bit better 27
127. s used experimentally for the HEARO RNA 15 with output in demo output svg HEARO svg or demo output pdf HEARO pdf 5 8 6 12 Drawing circles associated with loops draw_circ label label Draws full circle for layout involving the nucleotide at the given label This command is useful for internal loops terminal loops and bulges For multistem junctions use the draw_circ directive associated with the multistem_junction_circular_ commands These circles are sometimes useful in preparing other annotation in a general purpose drawing program 5 8 6 13 Drawing direct lines between consecutive nucleotides straight_line_3prime label label Draws the straight line connecting the center of one nucleotide to the next nucleotide 3 to it 5 8 7 Miscellaneous 5 8 7 1 Override default parameters for drawing SetDrawingParan The syntax for this command is the same as the SetDrawingParam command in the r2r_meta file except that names and parameters are space separated as are all R2R command parameters within the Stockholm file For details on parameters see Section 5 5 2 Note due to the way R2R processes files changing a drawing parameter in one line of a file might affect the use of that parameter in earlier lines of the file 5 8 7 2 No 5 label no5 does not draw the 5 label on the molecule Useful for drawing pseudoknots and some modular structures 70 5 8 7 3 Adding G for transcription g for transcripti
128. se This case arises with the first pseudoknot in SAM IV riboswitches 16 for which I disabled auto matic joining see demo SAM IV SAM IV sto 3 8 3 Pseudoknots drawn with in line style Sometimes the geometry of a particular RNA motif allows the pseudoknot base pairs to be drawn without the use of a callout In these cases it is often desirable to use an in line drawing of the pseudoknots as shown in Figure 3 17 However even if you want to draw an in line style I recommend starting with a basic callout style drawing because it s easier to understand the RNA and decide on a final layout Internal loops bulges terminal loops and multistem junctions are by default drawn along a circle However adding place_explicit commands that refer to positions within these loop regions will cause them to be drawn in a straight line In cases where you want a part of the original loop to be drawn along a circle you can use the bulge or bulge_flip commands In order to achieve an in line pseudoknot layout it is often necessary to flip stems using the f option to the place_explicit command Sometimes it can be confusing to get the proper stems flipped It is useful to note that R2R has default rules for layout When you use a place_explicit command your command will override default rules Specifically it will override any conflicting default rules To diagnose problems with flipped stems try using the mark flip R2R command which shows you i
129. servation of nucleotides nor in calculation of base pair covariation Similar semantics apply to 3 ends that consist entirely of gap characters The fragmentary mode was also applied to HEARO RNAs 15 in the case of the 3 most hairpin that is often missing presumably due to truncation events Fragmentary mode is enabled by adding the following line to a sto file GF FRAGMENTARY 1 4 3 Weighting sequences using your own algorithms In some cases the GSC algorithm might not be the way you want to weight sequences In general the GSC algorithm is appropriate because it corrects for the fact that some RNA sequences might be overrepresented in a database because biologists have decided to sequence the organisms that contain those RNAs frequently ahem E coli or RNA sequences might be overrepresented because they luckily for them happen to be used by a lineage of organisms that are highly successful for reasons unrelated to the RNA However these concerns might not hold in some situations To weight sequences yourself you must create a version of your Stockholm alignment that has a line specifying the weights as follows GF USE_THIS WEIGHT MAP map where map is a space separated list where each element of the list specifies a hitId identifying a sequence in the alignement followed by a space followed by the weight of that sequence For example the following is a valid Stockholm file with custom weights STOCKHOLM 1 0 A
130. shape Skele ton drawings can be applied to consensus or to individual sequences and are useful to give a compact summary of the whole structure Larger skeleton drawings are demonstrated in the files demo exceptional GOLLD sto and r2r_meta and demo exceptional HEARO sto and r2r_ meta Those files also demonstrate making scaled down thumbnail skeleton drawings A more realistic example of a skeleton drawing is given in Figure 3 4B Key points e Fields in the r2r_meta file are separated by tab characters e A single RNA molecule can be drawn using the oneseq directive in the r2r_meta file 16 Figure 3 3 demol iii Other kinds of drawings A c intermediate demo1 iii cons sto demo1 iii sto oneseq bigfoot intermediate demo1 iii cons sto skeleton with pairbonds demo1 iii gt Y magic G G C 0 0 B tab character p970 YGa AC STOCKHOLM 1 0 Zani human ACACGCGAAA GCGCAA CAAACGUGCACGG chimp GAAUGUGAAAAACACCA CUCUUGAGGACCU demo1 iii bigfoot bigfoot UUGAG UUCG CUCGUUUUCUCGAGUACAC A 7e GR bigfoot DEL_COLS sssssssssscceceeeee 2 GC SS_cons ELTE o ra GC R2R_LABEL SN A re ee IA Lsssss aE SA GC R2R_XLABEL_bf rss C R2R_XLABEL_bfl a a DESE A EEEE F R2R var_hairpin lt gt F R2R var_backbone_range 1 2 F R2R tick_label L magic G F R2R_oneseq bigfoot outline_nuc bf U F F F _ magic G R2R_oneseq bigfoot tick_label bf1 U bigfoot UNCG loop R2R if_skeleton SetDrawingParam
131. sizes note as with all font sizes default units are points for font sizes only nameFontSize for text that shows the name of RNA motifs Default 12pt nucFontSize writing nucleotides Default 7 5pt e Nucleotide layout distances internucleotideLen distance between consecutive nucleotides along the backbone Default 0 105in pairLinkDist distance between base paired nucleotides Default 0 17in e tick label nucleotide labels nucTickLabel_distFromNuc the distance from the nucleotide at which the tick line starts Default 3 75pt nucTickLabel tickLen the length of the tick line Default 4pt nucTickLabel tickPenWidth the width of the tick line Default 0 5pt 45 nucTickLabel_extraSpaceToText extra distance between the end of the tick line and the start of text might be a bit off because R2R still does not use absolutely exact font metrics Default 1 5pt nucTickLabel_fontSize the font size used to draw the label Default 6pt e backboneWidth width of lines that represent the backbone used for skeleton drawings and for variable length regions Default 1 5pt e outlining inlining nucleotides also used to indicate nucleotides involved in pseudoknot pair ings outlineNucExtraRadius extra distance from outside of nucleotide that is added before the outline line Default 3 75pt outlineNucPenWidth width of pen used to draw the outline Def
132. sto E another resolution break the pair in R2R making it a bulge file demo demo breakpair fix3 sto You can also make R2R apply this solution automatically to all otherwise broken base pairs by adding GF R2R SetDrawingParam autoBreakPairs true to the file This is useful when you want to draw an RNA quickly F another res olution keep both sides of the pair The nucleotide that would normally have been deleted is drawn with a circle that uses a gray line rather than the normal black line file demo demo breakpair fix4 sto G The drawing resulting from part F 19 Figure 3 6 demo breakpair 2 Another cause A B STOCKHOLM 1 0 ERROR there was a problem drawing motif human ACACGCGAAA GCGCAA CAAACGUGCACGG demo breakpair varlen One or more pairs chimp GAAUGUGAAAAACACCA CUCUUGAGGACCU is getting broken by one side getting bigfoot UUGAG UUCG CUCGUUUUCUCGAGUACAC deleted presumably because it s not con H GC SS cons lt lt lt SE IS Sn served while the other one stays H GC R2R_LABEL cceesceevceeveneess dh iss BBs Consider GF R2R var_backbone_range 1 2 using GF R2R keep allpairs Note another explanation is that you ve used GF R2R var_backbone_range on columns that includes a pair text columns in align ment 33 40 left keep right chuck left context nR nCnCnn Y text columns in alignment 34 39 left keep right chuck left context R nCnCnn Yn A input Stockholm file demo d
133. t a is associated with a specific position on the page and is also associated with either a line or an arc that intersects that position However the line or arc does not necessarily intersect a or a i 1 We wish to find connectors from a to a 1 for all i If a and a 1 are both associated with lines and if the lines are coincident i e the lines are an extension of one another then we connect a and a with that line Similarly if they are both associated with arcs that are coincident we connect the anchor points with that arc If a is associated with an arc and if the arc of a intersects the position of a then we connect a to a 1 using that arc In this case it does not matter whether a is associated with a line or an arc Analogous logic applies to the case where a is associated with an arc and a is associated with a line and the arc of a 1 intersects the position of as The arc of a might not intersect the position of a For example the multistem_junction_ circular_solver leads to bulges that are drawn on a circle but do not in general intersect the adjacent paired nucleotides In cases like this we declare a join error A join error denotes cases where R2R is not implemented with logic to join the anchor points nicely In the event of a join error a is connected to aj with a straight line In practice these straight lines are not very 86 problematic because join errors are relatively rare and becaus
134. t drawing unit posTo 23 species that position 23 is the nucleotide po sition within the second drawing unit whose nucleotide ranges are given default rule means that this is the default placement which can be overridden with a place_explicit command 45 1 0 90 defines the place_explicit style coordinates where 45 is the placement angle 1 0 is the relative units to travel and 90 is the relative angle of the next unit Thus this is analogous to the following place_explicit command place_explicit position 23 position 22 45 10 0 0 90 The text involvesCircularLayout true within a link signifies that the posFrom or the posTo drawing unit is an internal loop that will be laid out along a circle Links relating to actual place_explicit commands look like this 73 link posFrom 32 169 174 272 277 raw 27 33 110 116 posTo 36 181 185 198 202 raw 36 41 51 56 place_explicit demo 104 leu phe leader cons sto 67 o 0 0 180 The file name and line number of the place explicit command are given The absolute shift coordinates are not shown 5 10 5 Selecting the place explicit commands Eventually you will see lines like Calling PositionBackboneElement for this place explicit These introduce the place explicit style links that R2R selected to place the drawing units There is a special link called dummy rule to position first element that gives the position of the 5 most drawing unit 5 11 In
135. tSubFamilyFromStockholm pl adding a command to the Makefile and draw it adding a line to the r2r_meta file However you can disable this on a per predicate basis by adding the line GF Makefile disablepred predname 7 2 2 Explicit rules You can explicitly direct the creation of a new type of drawing within the r2r_meta file as follows This also allows you to define additional symbols for use with R2R s ifdef and related commands Note that all fields are space separated in this file but the spaces will be converted to tab characters for the r2r_meta file The file demo 104 crcB sto contains a particularly overwrought array of example usages of these directives 7 2 2 1 Oneseq GF Makefile oneseg Directs the script to make a oneseg mode output For example GF Makefile oneseq NC_003888 3 314159 2653589 define alt style 1 7 2 2 2 Skeleton GF Makefile skeleton Directs the script to make a skeleton or skeleton with pairbonds drawing As above additional define commands can be given 7 2 2 3 Defines for consensus diagrams GF Makefile define Directs the script to make a drawing of the consensus with additional define commands 7 2 2 4 Using define directive with modular structure GF Makefile pred predname Directs the script to draw a subfamily defined by predicate predname with additional define commands after predname The predicate predname must be defined earlier in the file 82 7 2 2 5 O
136. ta is read into data structures is in the function OneStockholm_try_Pass1 The function OneStockholm try implements passes 2 process commands that remove nucleotide positions e g var backbone range and 3 all other commands ParseSs cpp Functions related to parsing RNA secondary structures in bracket notation i e the SS_cons lines PdfGraphics cpp Drawing into PDF files Concrete implemntation of AdobeGraphics interface PdfGraphics h Header file for above PositionBackbone cpp Implements most functions for determining the position of nu cleotides within a drawing PositionBackbone MultiStemCircular cpp Implements various methods to draw multi stem junctions except solvers PositionBackbone MultiStemCircularSolver cpp Implements solvers for multistem junc tions R2R Utils cpp Utility functions related to parsing sto files and positioning nucleotides R2R cpp Contains main function with top level functions Sets up default values parses r2r_meta files dispatches to other code R2R h Declares data structures classes and functions used by multiple files RnaDrawer cpp Classes to actually draw RNA diagrams based on already determined nu cleotide positions Includes code to find paths for the shade_along_backbone command or skeleton mode drawings SelectSubFamilyFromStockholm pl Perl script to support extraction of a specified subset of sequences from a sto file for modular structures Chapter
137. tems If you set an angle that is not nicely compatible with the tangent of the circle where the stem needs to go things will look a bit bad or the computer might not be able to find a plausible solution strategy says how to try to fit the base outermost pair of the stem into the circle strategy 60 is ignored for the bulgecircley options but you must provide a legal option Valid options for strategy are as follows strategy 1 lowercase L means that the left nucleotide lies on the circle and the right nucleotide should go wherever it has to strategy r is the reverse strategy m means that the midpoint of the outer basepair is coincident with the circle strategy aa allows the computer to set the direction of the stem to any angle allowing the base of the stem to be flush with the circle This makes the circle look best but then the stems are not likely to be horizontal or vertical Note to ease the implementation you still have to supply some number of angle even though this number is ignored for strategy aa Also aa is not implemented for _solver except for the enclosing stem It could be implemented for any stem but there doesn t seem to be a point strategy A number from 0 1 can be specified as the strategy In this case the circle will intersect at this point which is a linear interpolation from zero left nuc to one right nuc Thus strategy 0 is equivalent to strategy 1 strategy
138. ten looks okay and isn t a problem for unpaired nucleotides but sometimes does not look good One workaround is to use the shade along backbone command then use a drawing program to convert to an outline For example in Adobe Illustrator the outline stroke command will do this conversion Note users who are interested in implementing a more general solution might wish to use the code in the 2geom subdirectory of the Inkscape source code This code can create shapes that are unions of simpler shapes and so it should be possible to create robust outlines of arbitrary regions I have not implemented this strategy in R2R however A complication is that it will be necessary for the user to specify whether they want the inside of an internal terminal loop to be a part of the shape or not In general I find it easier to use shade_along_backbone and then manipulate the shape in Adobe Illustrator 69 5 8 6 11 Shading the backbone in skeleton drawings shade backbone skeleton label color This command is only relevant to skeleton mode drawings It is similar to shade_along backbone but shades the backbone itself Normally the backbone is drawn in black Note that the use of different colors for the backbone in skeleton drawings often does not look good because the isolated base pairs are small circles It can help to change the drawing style using this command SetDrawingParam alongBackboneStyle 1 see Section 5 5 2 The feature i
139. teracting with R2R using scripts This section discusses feature that are useful if you want to generate R2R input by scripts or if you want to extract layout information from R2R 5 11 1 Generating R2R input using scripts Scripts can explicitly select a layout for R2R using the place_explicit set_dir pos0 bulge and bulge flip commands These commands are described elsewhere in this chapter and in the tutorial An example input file is in the directory demo pedagogical ScriptInputExample sto First set the orientation of the backbone at the 5 most position set_dir pos0 angle in degrees see section 5 8 2 1 for more on this command and also how to flip the drawing so that stems are drawn in a mirror image Then set the position of key nucleotides using the place_explicit command see section 5 8 2 6 It will probably be most convenient the orient nucleotides relative to the 5 most nucleotide and to refer to alignment positions by their numbers For example to set alignment position 7 note the 5 most column is numbered zero to coordinates 4 5 oriented at an angle of 75 degrees relative to the angle of the 5 most nucleotide use this R2R command place_explicit 7 pos0 00045 75 Note e distance units are in terms of the generally constant distance between nucleotides by default 0 105 inches See the internucleotideLen drawing parameter on page 45 Thus by default coordinates 4 5 in the above place explicit command co
140. tides in the loop so that s why there are 7 positions Then we define a predicate GYRA which sees if those 4 nucleotides match GYRA 6 2 2 2 Perl predicates Perl predicates allow you to use arbitrary perl expressions which are convenient for Boolean logic The syntax is H GF SUBFAM_PERL_PRED predicate name perl code Currently there aren t that many things you can do with the perl code without implementing additional capabilities within SelectSubFamilyFromStockholm p1 At the moment you just have the associative array predValue predValue predicate name evaluates to true or false where predicate name is the name of a predicate predicate name can refer to either a Regular expression or a Perl predicated but the referenced predicted must have been defined earlier in the sto file than when it is used otherwise its value will be undefined The perl code used in defining a Perl predicate should end with a return statement NOTE predicate names cannot contain an underscore _ The problem is that it would be ambiguous in the SUBFAM_pred_label substitution code For example GF SUBFAM_PERL_PRED other return predValue GYRA predValue GRRA Perl predicates also have access to any GS tags for the current hit in the map gsTag This feature is used for GOLLD see demo exceptional GOLLD sto to eliminate sequences with regions that could not be structurally aligned given the limits of other s
141. tion i 1 also intersect the circle 1 Yes 0 No In some cases for example some multistem junction solver results the ends of a single stranded junction are on a circle that does not intersect with the nucleotide at position 1 This issue only has implications if you want to trace along the backbone of the RNA R2R has an ad hoc way of solving this Is position actually a variable length backbone symbol 1 Yes 0 No Is position actually a variable length hairpin Is position actually a variable length stem Is position actually a variable length terminal loop 76 Chapter 6 Reference modular structures 6 1 Sub families You can display subsets of a family with re normalized conservation stats to e display the base pairs forming a pseudoknot separately e display distinct sub types of a motif like the GRRA GYRA subsets of GEMM 13 11 Doing this requires running additional commandes these commands could be automated using a Makefile Ignoring the issue of automation you use the script SelectSubFamilyFromStockholm pl to extract the subset based on some predicate then you calculate conservation stats from it then you run the generated sto file through R2R Examples are given in Section 3 7 6 2 Using SelectSubFamilyFromStockholm pl Often pseudoknots and alternate stems cannot be drawn without breaking the RNA backbone In this case you need to draw the pseudoknot as a separate structure
142. to see what labels are available add an R2R command that refers to an invalid label e g GF R2R tick_label INVALID dummystring R2R will give you an error message that lists the labels that are valid 5 7 1 Main labels The main label line is GC R2R_LABEL which has 1 symbol per column You can create multi symbol label for each column s label with GC R2R_LABEL_i for integers i 1 2 If there is a dot in a column nothing is added i e the dot is the empty string For example the following Stockholm file has three columns with the first two columns having the labels A and mul STOCKHOLM 1 0 x AAU y CAG Z GGC GC SS_cons lt gt GC R2R_LABEL am GC R2R_LABEL_1 u GC R2R_LABEL_2 1 5 7 2 Extra named label lines Sometimes having only one set of labels with the R2R_LABEL line is not enough You can add additional labels with names with GC R2R_XLABEL_name for some name name The name must not end with a number You can create multi symbol names by adding lines like GC R2R_LABEL_2 for integers 2 The name of the label is relevant when you want to refer to the label Referencing labels is explained below in Section 5 7 5 5 7 3 Sequence specific optional You can also have sequence specific labels in oneseq mode In this case all labels override anything that is specified in the global GC R2R_ LABEL stuff To specify sequence specific labels GR hitId R2R_L
143. to two units 43 5 4 Data types in R2R 5 4 1 hitId I define hitIds as strings in Stockholm format files that identify each sequence In Stockholm files the hitId begins a line and is followed by whitespace which are then followed by the RNA sequence with gap characters 5 4 2 Distances Distances within drawing commands are specified in internucleotide units One internucleotide unit is the distance between two consecutive nucleotides along the backbone and defaults to 0 105 inches 5 4 3 Angles Angles are always specified in degrees 5 4 4 Measurements length width size Some constant lengths used by R2R can be set to non default values using the SetDrawingParam command see Section 5 5 2 The default units are inches except for font sizes Font sizes are by default measured in points Suffixes can be used to specify other measurement units No whitespace is allowed between the number and the suffix The following measurements are equivalent e 1 default is inch e lin just makes the default explicit e 2 54cm e 25 4mm e 72pt 5 4 5 Colors Colors e g in the circle nuc command are specified as follows e RGB colors are written in the form rgb r g b where r g b written separated by commas are numbers from 0 to 255 9999 e Standard colors used in cleavage diagrams for in line probing experiments can also be used They are specified as cleavage unknown no data gray cleavage constitutive yel l
144. uch ups in Illustrator If the bulge is defined such that it doesn t use a circular bulge R2R may fail e backbonelen label length Sets the length of the variable length backbone identified by label For example these two commands would make sense in a given file GF R2R multistem_junction_bulgey J backbonelen A 8 GF R2R var_backbone_range A B for columns labeled A B and J 5 8 3 2 Multi stem junctions automatic circular layout multistem_junction_circular label see below multistem_junction_circular_solver label see below multistem_junction bulgecircley_solver label see below multistem_junction_bulgecircleynormals_solver label see below multistem_junction_bulgecircley_stemnormals_solver label see below These commands are demonstrated and explained at a high level in Section 3 6 1 or Section 3 6 3 The current section explains the details All of these commands attempt to find a circle on which to approximately position the nucleotides along a multistem junctions The parameter label must specify a column that is the left nucleotide of the base pair that encloses the multistem junction Warning these commands often don t work very well It will fairly often produce wacky layouts with nucleotides crossing which happens if it cannot solve the non linear program optimization problem In rare cases the program will crash Note the selection of a good starting point seems to be important for the solver to solve high
145. ue false R2R will report an error Default false e DNA if this variable has the value true R2R will treat the input under the assumption that it is a single stranded DNA molecule In this mode and U nucleotide in the input will be changed into a T Default false name and value pairs may be repeated multiple times Thus for example the following line is legal and sets 2 parameters at once SetDrawingParam varHairpinNumFakePairs 2 varTerminalLoopRadius 0 12 5 5 3 Oneseq mode If a line has three fields and the second field is equal to the literal string oneseq then the first field is the path to a Stockholm alignment file and the third field is a hitId Section 5 4 1 specifying a specific sequence within the alignment In this case R2R will draw that sequence It can optionally draw the sequence like a cleavage diagram for in line probing experiments Section 5 5 6 49 5 5 4 Skeleton mode If a line has two fields and the second field is equal to either the literal string skeleton or skeleton with pairbonds then the RNA is drawn in skeleton mode In this mode a black line traces the backbone Such drawings are often used to present a small sketch of an overall structure especially for larger RNAs Drawing of nucleotides nucleotide conservation length of variable length regions and covariation annotation are supressed The lines that indicate base pair bonds are supressed with skeleton but drawn with skeleton with
146. us mode is done in the r2r_meta file Section 5 5 Note these commands are equivalent to certain commands with defines as follows e R2R_consensus is equivalent to ifndef oneseq e R2R oneseq NC_003888 3 100 200 is equivalent to ifdefeq oneseq NC_003888 3 100 200 5 8 1 3 Other kinds of conditional commands If acommand begins with any of the following words the rest of the command is ignored in certain circumstances e if_skeleton the rest of the command is processed only if skeleton drawing mode has been enabled see Section 5 5 4 Equivalent to ifdef skeleton e if not_skeleton reverse of above Equivalent to ifndef skeleton e if_cfsqp the rest of the command is processed only if CFSQP is available to the program Equivalent to ifdef cfsqp e if_not_cfsqp reverse of above Equivalent to ifndef cfsqp 5 8 2 Turning and positioning 5 8 2 1 Set_dir The set_dir command is used to set the orientation of the 5 most nucleotide The orientation of other nucleotides are set either by R2R s default layout rules or by place_explicit commands set_dir pos0 angle Sets the orientation at the 5 most nucleotide to angle set_dir pos0 angle f Sets the orientation and causes stems to be flipped reflected along the axis of their middle This is caused by the extra f 55 5 8 2 2 Laying out arbitrary units like a bulge bulge label Position the structural element starting at label as a bulge Note implicitly the bu
147. ver will not work R2R will report an error e Option 2 Install CFSQP Advantage all functionality will work Disadvantage CFSQP must be requested from AEM Design which requires additional time CFSQP allows for all the automated layout schemes for multistem junctions However CF SQP must be requested from AEM Design see below and is not available for commercial purposes Unfortunately I am not permitted to distribute CFSQP with R2R 2 3 Make 3rd party tools 2 3 1 Infernal R2R uses the Squid library that was distributed with an older version of the Eddy Lab s Infernal software I have removed much of the code that is not used by R2R The remaining subset of Infernal is in the subdirectory NotByZasha infernal 0 7 The full Infernal package along with more recent releases is available from http infernal janelia org To build the squid part of Infernal run the following three commands cd NotByZasha infernal 0 7 configure make Note e No installation of Infernal is necessary R2R uses only the infernal 0 7 squid libsquid a file e You must use Infernal version 0 7 More recent versions of the Infernal software might not work and Infernal version 1 0 is guaranteed to fail 2 3 2 CFSQP optional If you want to use the CFSQP solver you must request the proprietary file cfsqp c Go to http www aemdesign com or http www ece umd edu Newsletter vol5 no1 fsqp htm or Google for CFSQP The contact
148. yout 12 e demo pedagogical Some drawings I did to explain issues in RNA drawing to non biologists or for the R2R paper Figures within each directory are organized by motif name Thus for example in Weinberg et al 2009 15 Figure 2a is a consensus diagram of GOLLD RNA This corresponds to the files demo exceptional GOLLD sto and demo exceptional GOLLD r2r_meta If you run r2r on these files the output will be found in output pdf GOLLD pdf or output svg GOLLD svg Similarly Supplementary Figure 1 of the same previous publication contains a skeleton diagram of GOLLD RNA Supplementary Figure 2 has an expanded consensus diagram and Supplementary Figure 3 holds a different skeleton diagram and a drawing of certain parts of the L brevis GOLLD RNA All these figures are generated from the same files and R2R output used for this will also appear in output pdf GOLLD pdf or output svg GOLLD svg However updates to the way that consensus nucleotides are calculated with degenerate IUPAC nucleotides have slightly changed the picture Versions of R2R up to 1 0 1 will exactly reproduce the old drawing 3 2 Example legend to help you with finished drawings To create a finished drawing for a published figure a legend is necessary to explain the annotations We have created generic annotations usable for a legend e Use the file demo Additional file 3 pdf for PDF format that can be imported into Adobe Illustrator or CorelDRAW e Use the file

R2R user manual - The Breaker Lab

Contents

Download Pdf Manuals

Related Search

Related Contents