TICO User's Guide
Contents
1. Simple coord files should be passed with the parameter co Note that either a file in simple coord format or in GLIMMER format may be given 7 1 3 Sequence format The genome sequence should be given in FASTA format as shown below The first line may con tain details for identification of the organism but may also be omitted As symbols for nucleotides both upper and lowercase characters are accepted To be processed the sequence should only con tain valid nucleotide symbols according to the IUPAC standard table of valid symbols is available on the TICO website For the training only the symbols A C G and T upper and lower case are considered All other IUPAC symbols will be ignored Example for the FASTA format 13 gt gi 6626251 gb U00096 1 U00096 Escherichia coli K 12 AGCTTTTCATTCTGACTGCAACGGGCAATATGTCTCTGTGTGGAT TAAAAAAAGAGTGTCTGATAGCAGC CTGAACTGGTTACCTGCCGTGAGTAAATTAAAA TATTGACTTAGGTCACTAAATACTTTAACCAA ATAGGCATAGCGCACAGACAGATAAAAAT TACAGAGTACACAACATCCATGAAACGCAT TAGCACCACC ATTACCACCACCATCACCAT TACCACAGGTAACGGT GCGGGCTGACGCGTACAGGAAACACAGAAAAAAG CCCGCACCTGACAGTGCGGGCT TTT TCGACCAAAGGTAACGAGGTAACAACCATGCGAGTGTTGAA GTTCGGCGGTACATCAGTGGCAAATGCAGAACGTTTTCTGCGTGTTGCCGATATTCTGGAAAGCAATGCC AGGCAGGGGCAGGTGGCCACCGTCCTCTCTGCCCCCGCCAAAAT CACCAACCACCTGGTGGCGATGATTG The FASTA file should be given with the parameter s2. 7 2 Output formats At the moment TICO provides three kinds of output format An adapted GLIMMER format general feature format GFF and the simple coord format The names of the output files are set in the configuration file of TICO If the name is omitted in the configuration file respective output will not be written The output directory of TICO in the default configuration will contain the files tico_results coord tico_results gff tico_results gl tico weights tis altsegq tis ref2line tis res tis seq The files tico_results coord tico_results gffandtico_results gl contain the pre diction of TICO as described in the following sections The files tis seq and tis altsegq con tain the sequence windows around the candidate TIS tis seq contains the candidates that were given to TICO during the invocation i e the candidates initially labeled as strong TIS tis altseq contains the additional candidates generated by TICO i e the candidates initially labeled as weak TIS In the file tis ref2line the mapping of the initially labeled weak candidates to the re spective strong candidate are denoted That is to say this file contains for each sequence from tis altsegq the line number of the respective sequence in tis seq If for example the first three candidates in tis altseq are denoted with the number 1 these three plus the first from tis seqare the TIS candidates for the first gene In the file tico weights the weights cal
3. 1 L 564 4 670315 0 19 11356 10646 2 L 711 13 285624 0 The GLIMMER like output is by default written to tico_results gl 7 2 2 General Feature Format GFF TICO provides output in general feature format GFF 4 The output is given according to the spec ifications at the Sanger Institute http 7 www sanger ac uk Software formats GFF GFF_Spec shtml lt seqname gt lt source gt lt feature gt lt start gt lt end gt lt score gt lt strand gt lt frame gt attributes comments By default the GFF output is written to a file named tico_results g Example output gff version 2 Type DNA Escherichia_coli glimmer tico CDS 337 2799 5 347931 Escherichia_coli glimmer tico CDS 2801 3733 11 448764 Escherichia_coli glimmer tico CDS 3734 5020 6 230648 Escherichia_coli glimmer tico CDS 5088 5237 3 815619 Escherichia_coli glimmer tico REANNCDS 5310 5741 0 111382 shift 21 note weak tis Escherichia_coli glimmer tico CDS 5310 5720 0 111382 note weak tis Escherichia_coli glimmer tico CDS 5683 6459 0 234908 Escherichia_coli glimmer tico CDS 6529 7959 19 75313 Escherichia_coli glimmer tico REANNCDS 8238 9191 19 169035 shift 63 Escherichia_coli glimmer tico CDS 8175 9191 19 169035 Escherichia_coli glimmer tico REANNCDS 9306 9893 19 613488 shift 3 Escherichia_coli glimmer tico CDS 9303 9893 19 613488 Escherichia_coli glimmer tico CDS 9928 10494 4 670315 Escheric
4. pected the first thing to do is to check the log file of TICO and if there is output in the command line The log file is located by default in var tmp tico log Windows c var tmp tico 1log There you find information of the settings for example paths and arguments and the steps per formed by TICO log level INFO Errors are denoted with the log level WARN ERROR or FATAL 8 1 Java Memory Error If you want to predict the TIS for a large genome the Java heap space may overflow The following exception is displayed in the commandline Exception in thread main java lang OutOfMemoryError Java heap space You should add the parameter Xmx to the Java call in the starter script or in your commandline call if you start the TICO main class directly Example java Xmx512m TiCo s seq fna g geneData In the example the java heap space is set to 512MB The heap should not exceed 1 4 of your physical memory maximum is 1GB 8 2 Java virtual machine errors The MATLAB libraries use the Java virtual machine Especially on 64 bit machines this may cause some incompatibilities an errors You will get messages like Error occurred during initialization of VM Unable to load native library 16 You should check the java you have installed on the system In some cases the error may be solved if you set the MATLAB variable MATLAB_JAVA to your system java 8 3 Running the MATLAB Compiler generated program If you got
5. MCR available under Linux To run a MATLAB compiler generated stand alone application the MCR MATLAB Component Runtime library archive should be available on your system If it is not yet installed please follow the instructions below The installation only should be performed one time for all MATLAB compiler generated applications you want to run on your system For Linux the installation just means to unpack the MCR an than set the LD_LIBRARY_PATH If you use the starter skript of TICO the path will be set automatically to the MCR in the TICO_HOME directory If you do not use the starter skript you should set the variable LD_LIBRARY_PATH manually In the following the MCR directory is addressed MCR_ROOT The terms arch and arch2 should be set to the systems architecture for example arch g1nx86 and arch2 i386 for 32 bit Intel machines or arch glnxa64 and arch2 amd64 for AMD 46 bit machines respectively Set the variable LD_LIBRARY_PATH to S MCR_ROOT lt ver gt runtime arch MCR_ROOT lt ver gt sys os arch MCR_ROOT lt ver gt bin arch S MCR_ROOT lt ver gt sys java jre S arch jre1 5 0 lib S arch2 native_threads MCR_ROOT lt ver gt sys java jre S arch jre1 5 0 lib S arch2 client MCR_ROOT lt ver gt sys java jre S arch jre1l 5 0 lib S arch2 The term lt ver gt should be replaced by the version number for example v7 4 bash example for Intel 32 bit machines MCR_ROOT
6. TICO2 1 complete package The tool also is provided separately as a tarball which should be unpacked in the TICO home directory or as independent complete package version The visualization of the positional weights see figure is realized in the form of a colored scheme The colors represent the level of the weights of a trinucleotide at the respective position High positive weights produce deep red areas in the plot high negative weights produce deep blue ones intermediate weights are represented by orange yellow and green areas A color scale displays the colors with the associated weight values The positions correspond to those in the extract window see 6 1 1 p 9 which are aligned to the position of the candidate TIS denoted at position 0 The positions with negative value indicate the upstream region the positions with positive value indicate the downstream region respec tively Note that the last two positions of the sequences are not considered in the evaluation for 17 they do not represent a trimer So if for example the default values are used for up and down stream extraction the example sequences output in tis seqand tis altseq have the length 60 with the TIS candidates at position 31 giving the first position the index 1 For the evaluation the trimer occurrences are counted for position 1 58 Additionally to exclude boundary effects the weights of the first and the last three positions are neglect The latter pos
7. an error message like Error while running run_clustering The external interface did not work properly Different sources of error are possible 1 You did not unzip install the MCR components 2 The MATLAB path LD_LIBRARY_PATH is not set correctly in the starter script or in your environment 3 The directory containing the script run_clustering is not included in the PATH or is not accessible 4 The directory run_clustering_mcr of the older version has not been removed during in stallation of a newer version run_clustering_mcr is created during the first invocation of the MATLAB components That causes problems when the MATLAB components are changed for example if you installed a newer version to the TICO home directory 5 The output directory is not readable or not writable 8 4 For TICO with Mail interface 1 If no mail is send The mail server may be set incorrectly in the configuration file 2 If no output is included to the mail Check the commandline parameters the files that shall be attached to the mail should be given as parameter with a name they should be given coord filename glimmer filename gff filename see 6 1 1 p 9 9 Visualization of the Weights since TICO2 1 Since version 2 1 the weights matrix calculated by TICO is written to the file tico weights in the output directory The matrix can be visualized with the MATLAB compiler generated tool Weight sVis which is provided in
8. in preparation and will be inte grated in the next release 7 1 Input formats For the moment only GLIMMER 3 format and our own format called simple coord are pro vided for the post processing of gene predictions 12 7 1 1 GLIMMER predictions The input file may contain the whole GLIMMER output but needed is only the section with the putative genes TICO searches the file for the line Putative Genes and reads all predicted ORFs from there on Note In GLIMMER output the genes are denoted exclusive stop Example for an input file in GLIMMER format Putative Genes 2 337 2796 1 L 2460 3 2801 3730 2 L 930 5 3734 5017 2 L 1284 6 5088 5234 3 L 147 Vote 8 5720 5313 3 L 408 DelayedBy 10 L 21 10 6459 5686 1 L 774 12 7959 6532 1 L 1428 4 8175 9188 3 L 1014 5 9303 9890 3 L 588 7 10494 9931 1 L 564 9 11356 10646 2 L 711 GLIMMER files should be passed with the parameter g 7 1 2 Simple coord format Simple coord format is as the name indicates a simple format containing only an id the coordi nates of the gene inclusive stop codon and the strand denoted in the following format gt id_posl_pos2_strand _score Example for the simple coord format gt 2_337_2799_ gt 3_2801_3733_ gt 5_3734_5020_ gt 6_5088_5237_ gt 8_5310_5720_ gt 10_5683_6459_ gt 12_6529_7959_ gt 14_8175_9191_ gt 15_9303_9893_ gt 17_9928_10494_ gt 19_10643_11356_
9. to start the Java main class directly Therefore the paths should be adjusted manually as described below see section 5 2 1 p 8 The call in principle complies with the call from the batch script The difference is that you should add the Java call Additionally you should start the class from the TICO home directory or pass the TICO home directory as first argument Example From the TICO home directory C program files tico gt java TiCo s seq fna g glimmer out From anywhere C gt java c program files tico TiCo s seq fna g glimmer out 5 2 1 Adjusting the paths Under Windows the PATH and the CLASSPATH can be adjusted through the graphical interface of the environment settings or temporarily through the DOs box To alter environment vari ables permanently under Windows 3 x Windows9x or WindowsME you should set them in the file autoexec bat which is located in the root directory on your boot device probably G autoexec bat The syntax is the same as in the DOS box see below To set a path the clicking way you should find in the system settings start menu settings the point system One of the tabs is denoted advanced there you find a button environment variables A variable PATH is probably already defined as system variable In oder to avoid overwriting the system variable when you add a user defined variable path you should include the system variable denoted as PATHS Example Add a variable
10. Commandline parameters o cros e K 4 ei ee PLY EDEN ES ES ES GLA Configuration ile Cig 63 44 28 eae Sd ex paw PAPER ES EES G2 Conger Tie Loerie lt s be iier d BR a RR EWES HES ESSER SS ES 7 Input and Output formats PA Tapulformafs see e Gs to Sat See e eh CRE RE PERS BES EAN ET ee ee 711 GLUMMER predictions so cue T sae Wi R OE R R ee EEE ROE OA E Simple corde Ofna es fe BERG EER ERY CREE SEES ED ALs s T Toat oo ea he 2S Eee RE Se Ree PASSER RS 7A SOA oae he he ROS OE BEES ee eh ee SE EER RS fous Adapted GLIMMER format oe esoe eG eh ec hee Pee Be RSS 7 2 2 General Feature Format GFF ec p sey de eee ere de be el ele eas few Wome COOr T seve Ped soe we eK ee ae RE ew Ee eS 8 Troubleshooting ol Javea Memory Hier vccceleagiee eh ev eeetecatet cadet eakees 8 2 Java virtual machine errors a ee ee 8 3 Running the MATLAB Compiler generated program sso 8 4 For TICO with Mail interface oaoa a 9 Visualization of the Weights since TICO2 1 10 License terms 11 Links 11 12 12 12 13 13 13 14 14 15 16 16 16 16 17 17 17 19 20 Preface The readme of TICO2 1 is still in a state of revision therefor parts may be written unclear or con tain mistakes Unquestionable it needs to be improved The reader may excuse the present in sufficiency We are grateful for comments and suggestions concerning both the readme and the tool 1 Introduction TICO is a tool for post processing of gene pr
11. HOME MCR export LD_LIBRARY_PATH MCR_ROOT v74 runtime glnx86 S MCR_ROOT v74 sys os g1nx86 MCR_ROOT v74 bin ginx86 MCR_ROOT v74 sys Jjava jre ginx86 jrel 5 0 lib i386 native_threads S MCR_ROOT v74 sys java jre ginx86 jrel 5 0 lib i386 client S MCR_ROOT v74 sys java jre ginx86 jre1 5 0 1ib i386 The LD_LIBRARY_PATH should be available on your system thus it should be set in the shell e g inthe profileor bashrc if you want it to be avaliable permanently Please note the licence file for the MATLAB compiled application 3 Running TICO ona Linux platform In the following TICO_HOME refers to the directory in which TICO is installed You have two possi bilities to start TICO You can start the program with the bash script t ico or you can start the Java main class If you directly start the main class you need to perform the invocation in the directory TICO_HOME or pass TICO_HOME in the commandline Additionally it may be necessary to include TICO_HOME in the PATH of your system if is not included and TICO_HOME tico jarin your CLASSPATH The easier way is to start TICO with the starter script tico The script will start the Java class and pass over the TICO_HOME directory The PATH the LD_LIBRARY_PATH only in the com plete package version and the CLASSPATH will temporarily be adapted to include TICO_HOME MCR_HOME only in the complete package version and the jars respectively If you do no
12. Meinicke M Tech B Morgenstern and R Merkl Oligo kernels for datamining on biological sequences A case study on prokaryotic translation initiation sites BMC Bioinformatics 5 169 2004 3 A L Delcher D Harmon S Kasif O White and S L Salzberg Improved microbial gene identification with GLIMMER Nucleic Acids Res 27 23 4636 4641 1999 4 K Rutherford J Parkhill J Crook T Horsnell P Rice M A Rajandream and B Barrell Artemis sequence visualisation and annotation Bioinformatics 16 10 944 945 2000 20
13. TICO User s Guide Release 2 1 Maike Tech 8th October 2006 GACCGTAGAAGCCACAAAAAATGAATGTTAATTACCTGA CTGCAAGGACTGGATATGCTGATTCTTATTTCACCTGAATGCGCTTAT CCT TCCCACTAAGAGCTTATGAAATCCGTTTTTACGATTTCCGCCAGC TTTAAATAAAATTGCTGTCAATTTTACGTCTTIGTCCTGCCACATTCTT AAT TAGGGAGACGT TTAGATGGGTAAAATAAT TGGTATCGACCTGGGT CAATTCTAGGAAGGTTCCTCTCCGCCCGTGCATTCAGGCT TAAAAAAG TTACGCCGATATGATTTAAGT CGTGCCGATGAATTACTCGATAACTGG TCACCTGAAAGAGAAATAAAAAGTGAAACATCTGGATTCTTTAGCAGT AGCCATAAACGGCTCCCTTTTCATTGTTAGAGAGAAATGAGCACGTCT TCGTGAGTCATCGT CGCGCGGAAAT GACAGTGGGGC Contents 1 Introduction 2 Installation Linux Poke QUACK SIA oo econ age de Pi ee RM Pd dG we weld Pee he eR 2 2 Installation of the Java components c0 20 ae es ee BR a eee SO Pool Seling TRC CLASSPATH 62S ete sees se SES GHEE CHASE Hes 20 Making the MCR available under linux gt o cac 2464445454 eee Ee be oe 3 Running TICO on a Linux platform 31 Using thebash sipt o tek croce Fh EE Oe RE RS RO SPEED Oe ee os 3 2 Running the Java main class of TICO ss sc ew ey oe ee ye ee RE 4 Installation Windows 4l erre Guide oda eo ew eR ee ee eee a ew ew Be 5 Running TICO on a Windows platform 51 Starting TICO withthe batchascript s lt s ke ee ee ee ya ni ENEE OS 5 2 Running the Java main class of TICO under Windows 521 Adjusting SPAS ve oS ER Hee ee Cees a He eee RSS 6 Configuration of TICO 6 1 Commandline Parameters and Configuration File aaaea 61 1
14. ase note the license terms see section 10 p 19 3 2 Installation Linux In the following a description of the necessary steps to install and run TICO under Linux is given The Quick Guide gives a step by step instruction the individual steps are detailed in the suc ceeding sections 2 1 Quick Guide For both versions Java 1 5 needs to be available on the system For the complete package Unpack the tarball and start TICO with the starter script see 3 1 p 6 or adjust the PATH the LD_LIBRARY_PATH see 2 3 p 5 and the CLASSPATH see 2 2 1 p 5 and start the Java main class of TICO see 3 2 p 6 For the individual components 1 Make the Java components jars available on your system see 2 2 p 4 2 Install the MCR MATLAB Component Runtime library archive see 2 3 p 5 3 Setting the LD_LIBRARY_PATH needed for the MATLAB components see 2 3 p 5 4 Adding the necessary jars to the CLASSPATH see 2 2 1 p 5 5 Start TICO with the starter script see 3 1 p 6 or adjust the PATH and start the Java main class of TICO see 3 2 p 6 2 2 Installation of the Java components The Java Runtime Environment at least version 1 5 0 should be available on your system The Java API can be downloaded from the web interface of SUN http java sun com TICO addi tionally uses the following packages e Log4j log4j jar Can be downloaded from the Log4j home at http logging apache org log4j Java Acti
15. automated way In the default configuration the flag is set to 1 i e true so the sigma is automatically adapted The final sigma value is given in the last line of the file tis res in the output directory and in the log file If the flag is set to 0 the initial sigma value as set in the configuration file or given as commandline parameter is used as smoothing parameter The ROC flag can be either set in the configuration file or as commandline parameter Parameter roc commandline ROC config file Range 1 true or 0 false Default 1 Minimum gene length Specifies the minimum length gene after may have after reannotation of the TIS denoted in bp If the distance of a potential candidate TIS falls below the minimum length it is omitted from the list of candidates Parameter minlength commandline MinLength config file Default 60 bp Output directory If an output path is given in the commandline all files generated by TICO see 7 2 p 14 are written directly to this directory If no path is given the results are written to a new created folder in the default path which is set in the configuration file The preset default path is var tmp under Windows the default path is c var tmp The name of the output directory will be generated random numbers like 9104113053744858521 Parameter io commandline Out put Path config file 10 Summary of commandline parameters Required Parameters s seq file
16. called path case independent with the value PATH c program files tico assuming TICO is located in the directory c program files tico and a variable classpath with the value CLASSPATH c program files tico log4j jar c program files tico tico jar The dots in the example indicate that there is no newline To set a path or any other variable in the DOS environment you should use the command set like is shown in the example below Calling set without any parameters will display all environ ment variables Settings altered in the DOS box are only available in the respective shell and do only persist as long as the shell exist After closing the window or in a new command window the system settings are restored Example Setting the PATH and the CLASSPATH in the DOS box C gt set PATH SPATH c program files tico C gt set CLASSPATH SCLASSPATH c program files tico log4j jar verbt c program files tico tico jar Note that every component of the paths should be separated with a colon The dots in the example indicate that there is no newline 6 Configuration of TICO 6 1 Commandline Parameters and Configuration File In this section the handling and configuration of TICO are described Some parameter can only be set in the configuration file some may also be given in the commandline If a parameter is 8 given in the commandline the respective value from the configuration file will be overw
17. culated by TICO during the clustering are written the file tis res contains the results of the clustering algorithm Each line contains the label or the line number and the PWM score calculated for the candidate TIS 7 2 1 Adapted GLIMMER format The output is denoted in a GLIMMER like format That means it contains all predictions from the input file in the same format like GLIMMER with two additional columns from the TICO prediction In the first column after the GLIMMER output the PWM score is given in the second the shift of the start during reannotation Additionally genes with a negative score are labeled with a hash mark at the end of line See also section 7 1 1 p 13 lt id gt lt start gt lt stop gt comments lt PWM score gt lt shift gt 14 The shift is given in respect of the strand of the gene A positive value means the reannotated start is located upstream of the original start a negative value indicates a downstream shift If the value of the shift is 0 the start is not changed from the original prediction Example output Putative genes 2 337 2796 1 L 2460 5 347931 0 3 2801 3730 2 L 930 11 448764 0 5 3734 5017 2 L 1284 6 230648 0 6 5088 5234 3 L 147 Vote 3 815619 0 8 5741 5313 3 L 408 DelayedBy 10 L 21 0 111382 21 10 6459 5686 1 L 774 0 234908 0 12 7959 6532 1 L 1428 19 753130 0 14 8238 9188 3 L 1014 19 169035 63 t9 9306 9890 3 L 588 19 613488 3 17 10494 9931
18. edictions with regard to improve the prediction of the correct translation initiation sites TIS Therefor TICO requires the input of a sequence in FASTA format and a set of predicted genes With the initially predicted TIS TICO generates a set of candidates and scores each candidate in an iterative process by means of a clustering algorithm The initially predicted TIS will be relocated if another candidate belonging to the respective ORF has a higher positive score A description of the algorithm can be found at the TICO web interface http tico gobics de and in 1 Since version 2 0 TICO has the feature of automated adaption optional of the smoothing pa rameter sigma see section 6 p 8 Also a bug has been eliminated causing an ArrayIndexOutOf BoundException under some circumstances very rare Since version 2 1 the weights calculated by TICO can be visualized with the tool Weight sVis as detailed in 9 p 17 There are some requirements to be fulfilled to run TICO properly Please read the readme file carefully at least part of it before you try to run the program If you did not succeed to install and run the stand alone version of TICO find any bugs or have any further questions please send an email to the following address tico info gobics de The easiest way to install TICO is from the complete package version Than you only have to have Javal 5 available on your system and to install the MATLAB libraries only for Windo
19. ghts 18 TICO Weights ACA AGA ATA CAA CCA Oligonucleotide QO Q O 2882298 gt bP FY gt gt 3 TCA TGA TTA 27 2 3 Position Figure 1 Visualization of the weights matrix calculated for E coli using the default settings 10 License terms The complete package is free for academic use The MATLAB compiler generated files are dis tributed under the MathWorks Inc Software License which is included in the package The Java classes of the user interface will be OpenSource when the code is cleaned and com pletely documented 19 11 Links TICO Java Others TICO home http tico gobics de G ttingen Bioinformatics http gobics de University of Gottingen http 7 www uni goettingen de Sun Java Home http java sun com Java 1 5 Documentation http java sun com j2se 1 5 0 docs api Java Activation http java sun com products javabeans glasgow jaf html Java Mail http java sun com products javamail Log Home http logging apache org log4j MathWorks Home http www mathworks com The Sanger Institute http www sanger ac uk GFF specification http www sanger ac uk Software formats GFF GFF_Spec shtml GLIMMER Home http www tigr org software glimmer References 1 M Tech N Pfeifer B Morgenstern and P Meinicke Tico A tool for improving predictions of prokaryotic translation initiation sites Bioinformatics 2005 2 P
20. he TICO home directory The handling of the logging is commented in the configuration file and is documented in detail in the Log4j documentation which is available at Log4j home see 11 p 20 Examples for the default the logging format 2005 03 04 17 06 22 817 ERROR GlimmerOutputParser Not a valid Glimmer output file Line 3 2005 05 07 14 46 54 522 ERROR TiCo Error while running run_clustering The external interface did not work properly IO error reading results no results were written Please consult the readme file By default TICO logs to the file var tmp tico log If you would like to log to another path you should modify the line log4j appender R File var tmp tico log in the log4j configuration file If you want the logging to be directed to standardout the line log4j rootLogger INFO stdout R should be commented in and the line log4j rootLogger INFO R should be commented out As you see the log level is set to INFO by default This is a verbose level TICO informs you which paths are set and which steps are performed The log level can be changed by replacing the keyword INFO with WARN ERROR or FATAL But note that even an event reported in the level WARN may cause the tool to produce empty results 7 Input and Output formats In this section the input and output formats are described which are provided by TICO at present Additional input and output formats e g GenBank format are
21. hia_coli glimmer tico CDS 10643 11356 13 285624 The GFF output can be visualized with the program ARTEMIS 4 The originally predicted genes are denoted with the tag CDS By default the entries with the tag CDS are displayed in light blue The relocated TIS are denoted with the tag REANNCDS To visualize this entries the tag should be added to the configuration file by default ARTEMIS_HOME etc options of ARTEMIS for example by adding the line colour_o_REANNCDS 1 Through this adaption the relocated TIS will appear grey 15 7 2 3 Simple coord format By default output in simple coord format is written to a file named tico_results coord The format is denoted as shown below See also section 7 1 2 p 13 Example output gt 2_337_2799_ _5 347931 gt 3_2801_3733_ _11 448764 gt 5_3734_5020_ _6 230648 gt 6_5088_5237_ _3 815619 gt 8_5310_5741_ _ 0 111382 gt 10_5683_6459_ _0 234908 gt 12_6529_7959_ _19 753130 gt 14_8238_9191_ _19 169035 gt 15_9306_9893_ _19 613488 gt 17_9928_10494_ _4 670315 gt 19_10643_11356_ _13 285624 8 Troubleshooting In this section some difficulties are summarized that may occur during installation and invocation of TICO This chapter will be extended when new sources of error are reported by users of TICO So do not hesitate to give us feedback if you had problems running the tool If you start the program and some error occur for example you do not get the results you ex
22. indow is used for the unsupervised learning It is assumed to contain the characteristics of respective start site e g the ribosome binding site e up Parameter exu commandline Ext ract Up config file Specifies the number of nucleotides to be be extracted upstream 5 a given start posi tion Default 30 nucleotides Minimum 10 nucleotides Maximum 100 nucleotides e down Parameter exd commandline Ext ract Down config file Specifies th number of nucleotides to be extracted downstream 3 a given start posi tion inclusive start Default 30 nucleotides Minimum 10 nucleotides Maximum 100 nucleotides Sigma The standard deviation parameter sigma of the Gaussian density specifies the smoothing 2 of the positional probabilities of the second order Markov Models A high value for sigma means the positional probabilities are highly smoothed The parameter doesn t imply any assumptions on trinucleotide positions in the sequence but adapts the estimation to a varying number of genes under consideration The default value 0 5 works well with approximately 4000 genes For a set with a smaller number of genes it may be useful to chose a higher value for sigma to prevent vanishing probabilities Parameter sig commandline Sigma config file Range 0 1 2 0 Default 0 5 ROC flag since release 2 0 The ROC Receiver Operating Characteristics curve can be used to optimize the smoothing parameter sigma in an
23. itions are part of the weights matrix but are excluded from the visualization So the visualization of a weights matrix calculated for the settings described above would show the position 27 0 24 WeightsVis can be started from a script adapting the necessary paths and setting some de fault parameters or the program may be called directly During invocation a flag with value 0 or 1 may be given to the script The default value of the flag is 0 If the flag has value 1 the trinucleotide with maximum positive score is marked in the visualized matrix Additionally the title and the labels of the axis may be altered in the starter script or during the direct invocation of the program Syntax for invocation with the starter script vis weights fil max flag Example for invocation with the bash script vis var tmp 5793472968765 tico weights Example for invocation with the batch script vis c my documents 758943764892 tico weights Note for invocation with the batch script The maxF lag cannot be given in the commandline but may be altered in the batch script vis bat Syntax for invocation without the starter script WeightsVis oligos figure title X Label Y Label weights file max flag Example for invocation without the starter script WeightsVis oligos TICO Weights Position Trinucleotide 123 tico weights 0 Example for invocation with the batch script vis bat c var tmp 5793472968765 tico wei
24. o the log file If you would like use the mailer interface your outgoing mail server should be entered key mail smtp host You also should store the email address of the sender in the parameter fromAdr The recipients address may also be set in the configuration file DefaultRecipient or can be given in the commandline with the parameter u Change the Nucleotide or Codon Table You can change the nucleotide symbols to be con sidered by TICO as well as the sets of start and stop codons Therefor you should add adjust the respective table files to the TICO home directory Example files are given with in the TICO tar ball TICO_HOME nucleotides and TICO_HOME codons The name of the files which should be read by TICO should be given in the configuration file with the key Nucleotides or CodonTable respectively By default without including a codon table ATG GTG and TTG are considered as potential start codons and TAG TAA and TGA as stops To change the set of start or the stop codons first of all you should set a codon file in the config of TICO by default TICO_HOME tico Then you should include exclude codons as given in the example 11 STOPCODON TAG STOPCODON TGA STOPCODON TAA this stop codon is excluded STARTCODON ATG You can either delete the respective line or just comment the line out 6 2 Configure the Logging The logging can be adapted with the configuration file 1og4j which is located in t
25. ritten All commandline parameters are summarized in table 6 1 1 6 1 1 Commandline parameters Search Range Specifies the range to be searched around putative gene starts for alternative start sites I e by the search range the maximum distance to a predicted TIS as derived from the input file is defined In this range all potential start sites are considered as candidate TIS A potential start site is defined as start codon that shares the same reading frame of the respective gene with no inframe stop codon between the start codon and the annotated stop At first the initially predicted TIS is labeled as strong TIS the alternative start sites are labeled as weak TIS During the iterative classification the label strong is assigned to the candidate start with the highest PWM Score i e the value from the positional weight matrix among the candidates of a TIS e up Parameter su commandline SearchUp config file Specifies the maximal distance to a given start position for upstream 5 alternative starts Default 250 nucleotides Minimum 50 nucleotides Maximum 250 nucleotides e down Parameter sd commandline SearchDown config file Specifies the maximal distance to a given start position for downstream 3 alternative starts Default 250 nucleotides Minimum 50 nucleotides Maximum 500 nucleotides Extract Range Specifies the range to be extracted around each candidate start site The resulting sequence w
26. rter script see 5 1 p 7 or adjust the PATH the CLASSPATH see 5 2 1 p 8 and start the Java main class of TICO 5 2 p 7 5 Running TICO on a Windows platform To run TICO under Windows you need to use the Windows commandline interpreter or DOS box as it is also called A GUI for TICO will be provided in the next version In the examples the DOS prompt is denoted like C gt where means that you are in the root directory of device C If an error occurred during the invocation or while TICO is running check the log files for error messages STEMP tico 1log 5 1 Starting TICO with the batch script To run the starter script you should type the path relative or absolute of the batch script and pass the sequence file the gene data file and other parameters like is shown in the example Note that you should set the path in double quotes if it contains any delimiter characters like whitespace Also note that the files should be given with the absolute path Example C gt c program files tico tico s c data seq fna g c data glimmer out In the example TICO is located in C program files tico The batch file will temporar ily adjust the PATH and the CLASSPATH to include the program run_clustering and all neces sary jars 5 2 Running the Java main class of TICO under Windows If the batch file does not work on you Windows maybe the DOS version does not support all necessary commands you have the possibility
27. t use a bash it may be necessary to adapt the script at least the interpreter call 3 1 Using the bash script If you call TICO with the starter script you only need to pass over the sequence file and a file containing the predicted genes It is necessary that the script and the configuration file are located in TICO_HOME tico s seq fna g glimmer out 3 2 Running the Java main class of TICO If you would like to run the Java main class TICO_HOME should be available in the PATH of your system and tico jar as well as the other jars should be included in the CLASSPATH That is nec essary for the MATLAB components and the Java classes to be found Furthermore it is necessary to pass TICO_HOME during invocation if you do not perform the invocation from TICO_HOME Otherwise the configuration file will not be found Example java TiCo STICO_HOME s seq fna co coord out If an error occurred during the invocation or while TICO is running check the log files for error messages default var tmp tico log 4 Installation Windows The necessary adaption to install and run TICO properly under Windows are enumerated step by step in the Quick Guide A detailed description of the steps to perform can be found in the referred sections 4 1 Quick Guide Java 1 5 needs to be available on the system see 2 2 p 4 1 Install the MCR MATLAB Component Runtime library archive with MCRInstaller exe 2 Start TICO with the sta
28. the sequence in FASTA format see section 7 1 3 p 13 and one of the following files containing the initial prediction g glimmer file GLIMMER predictions see section 7 1 1 p 13 co simple coord file predictions in simple coord format see section 7 1 2 p 13 Optional Parameters io dir output directory default range default value su number upstream search window 1 500 250 sd number downstream search window 1 500 250 exu number upstream extracted window 1 100 30 exd number downstream extracted window 1 100 30 minlength number minimum gene length 10 30 sig number smoothing parameter 0 1 2 0 0 5 roc 1 0 ROC flag 1 or U 1 TICO with mail interface u user domain xy Email address glimmer file name GLIMMER output will be attached see section 7 2 1 p 14 coord file name output in simple coord format will be attached see section 7 2 3 p 16 gff file name output in GFF will be attached see section 7 2 2 p 15 6 1 2 Configuration file tico The configuration file provides the possibility to configure the external interfaces and to set de fault values to shorten the commandline call In the package an example tico is given with comments on the parameters To leave a parameter empty modifies the behavior of TICO in some cases In the case of the output files the result in respective format is not written in consequence of a deleted value If a key is missing an entry of level WARN is written t
29. vation and Java Mail are only needed for the version with mail interface e Java Activation Framework activation jar Available at http java sun com products javabeans glasgow jaf html e Java Mail mail jar Can be downloaded from SUN at http java sun com products javamail Theses jars are included in the complete package version and will be included temporarily in the CLASSPATH during the invocation of TICO with the starter script If you do not use this version the jars should be downloaded and installed separately additionally they should be included in the CLASSPATH see next section 2 2 1 Setting the CLASSPATH only needed if you do not use the starter script and the complete package version Your CLASSPATH should include the jar Java archive of Log4j and if you use the TICO version with mail interface you additionally should include Java Activation Framework and Java Mail Example without mail interface export CLASSPATH usr share java log4j 1 3alpha 3 jar Example with mail interface export CLASSPATH usr share java mail jar usr share java activation jar usr share java log4j 1 3alpha 3 jar Note The jar of TICO TICO_HOME tico jar with TICO_HOME referring to the directory in which TICO is installed should be included to the CLASSPATH if you do not use the starter script to run TICO If you use the script the PATH and the CLASSPATH will be temporarily extended see section 3 1 p 6 2 3 Making the
30. ws However if for any reason you prefer to install the components to separated locations or if some of the components already are provided on your system the necessary steps under Linux are detailed in this readme Currently two commandline versions of TICO for Linux and Windows are provided The ver sions differ in the mail interface which is only provided by one of the implementations This version of TICO requires Java Mail an Java Activation to be available on your system If it is not necessary to send the results of TICO anywhere via mail the TICO version without mail interface would be sufficient The implementation of TICO consists of two parts The user interface have been implemented in Javal 5 and requires the Java Runtime Environment 1 5 JRE1 5 The clustering routines were developed under MATLAB and were compiled with the MATLAB compiler The tool is developed and tested on Debian Linux Sarge WindowsXP and Windows2000 The jar file t ico jar is platform independent as well as the additionally needed jars while the starter script and the MATLAB classes are platform dependent The Linux examples are given for a bash and the starter script is also written for a bash If you use another shell some modifi cations may be necessary The batch script to start the main program under Windows is written and tested for DOS version 7 0 and higher If you use an older DOS version you should start the TICO main class manually Ple
Download Pdf Manuals
Related Search