Assignment 2 - Sequence-based predictions - Ping-Pong
Contents
1. gt sp P07024 USHA_ECOLI Protein UshA OS Escherichia coli strain K12 GN ushA PE 1 SV 2 MKLLORGVALALLTTFTLASETALAYEQDKTYKITVLHTNDHHGHFWRNEYGEYGLAAQK TLVDGIRKEVAAEGGSVLLLSGGDINTGVPESDLOQDAEPDFERGMNLVGYDAMATGNHEFD NPLTVLROOEKWAKFPLLSANIYQKSTGERLFKPWALFKRODLKIAVIGLTTDDTAKIGN PEYF TDIEFRKPADEAKLVIQELQOTEKPDI I IAATHMGHYDNGEHGSNAPGDVE RAL PAGS LAMIVGGHSQDPVCMAAENKKOVDYVPGTPCKPDOONGIWIVQAHEWGKYVGRADF EFRNGEMKMVNYQLIPVNLKKKVTIWEDGKSERVLYTPEIAENQQMISLLSPFONKGKAOL EVKIGETNGRLEGDRDKVRFVOTNMGRLILAAQMDRTGADFAVMSGGGIRDSIEAGDISY KNVLKVOPFGNVVVYADMTGKEVIDYLTAVAOQMKPDSGAYPQFANVSF VAKDGKLNDLKI KGEPVDPAKTYRMATLNFNATGGDGYPRLDNKPGYVNTGF IDAEVLKAYIQKSSPLDVSV YEPKGEVSWQ SignalP NN prediction gram networks Sequence MKLLQRGYALALLTTFTLASETALAYEQDKTYKI TY YLHTNDHHGHFWRNE YGEYGLAAQKTLYDGIRKE Position gt Sequence length 70 Measure Position Value Cutoff signal peptide max Y 26 0 693 O33 YES max S 10 0 920 0 92 YES mean S 1 25 0 607 0 49 YES D 1 25 0 650 0 44 YES Most likely cleavage site between pos 25 and 26 ALA YE The signal sequence is between positions 1 25 whereas the cutoff cleavage site is at position 26 probability 0 940 Name Len mTP SP other Loc RC Sequence 550 0 055 0 044 s 1 cutoff 0 000 0 000 0 000 The sequence is2. amino acid residue types for individual positions However while profiles are an empirical attempt to generalize BLAST like alignment scoring to individual scoring of each alignment position HMMs have a much more solid base in mathematical statistics and are therefore more reliable The figure below shows a very simple HMM which is very short and also uses DNA simply because DNA has fewer letters than proteins Insert state C Match state Delete state 2 p OC N NH gt ie e gt Te gt 4 ul O In HMMs everything is a statistically computed probability The profile scores have been replaced by probabilities the general insertion and deletion penalties from profiles have been replaced by specific transition probabilities between states there are probability distributions for the types of amino acid residue present in insertions and so on All this makes the HMMs very sensitive and powerful and they can therefore be used to model very large and widely diverse groups of sequences such as for example protein superfamilies A less simplified illustration of an HMM matching a protein family is available here Less simplified HMM example SAM drawmodel format pdf Pfam is a database of high quality HMMs designed to reliably identify protein domains Go to Pfam and look around Note that Erik Sonnhammer Stockholm Bioinformatics Center is one of the guys behind this important
3. as repellent signals during axon and viral proteins guidance Act as repulsive axon guidance cues during development or involved in immune function Plexin repeat Description Plexins receptors for multiple A cysteine rich repeat found in several classes of semaphorins different extracellular receptors IPT TIG domain Description NA cell surface receptors intracellular transcription factors Involved in DNA binding Protein tyrosine kinase Description Eukaryotic protein kinases are Protein kinases are a group of enzymes that enzymes belonging to an possess a catalytic subunit which transfers extensive family of proteins It the gamma phosphate from ATP to one or share a conserved catalytic core more amino acid residues in a protein common to both substrate side chain resulting in a serine threonine and tyrosine conformational change affecting protein protein kinases It is involved in function ATP binding catalytic activity of the enzyme With this we can say that RON_HUMAN is a macrophage stimulating protein which is localized on the cell membrane It has ATP binding activity and has SEMA and IPT TIG domains One way of studying protein similarities is to search for homologues in sequence databases using for example BLAST compare exercise 1 Another way is to use databases that contain information on sequence patterns and protein family conservation such as Prosite and Pfam
4. database See the HELP pages and use Pfam to scan RON_HUMAN for matching HMMs Note In addition to its own HMMs Pfam also uses external tools to classify the query protein e g SMART seg signalp etc For this assignment however we will ignore all hits that do not come from Pfam itself 2 12 Does Pfam find any sequence features that Prosite did not If so which A Plexin and IPT TIG repeats were found 2 13 Did Prosite find any sequence features that Pfam does not If so which A No or Protein Kinase ATP binding domain with a more sensitive setting Clicking on a Pfam hit will take you to the corresponding documentation page which holds both biological and technical information on the hit For your amusement you can click on the link Download HMM far down to the right to see what a HMM really looks like on the inside 2 14 Using Prosite and Pfam try to find as much information on RON_HUMAN as possible What do you think this protein does in the cell Which parts do what Please give your best theories on the protein s function localisation interactions etc A Prosite Pfam RON_HUMAN Description Macrophage stimulating protein Macrophage stimulating protein receptor receptor alpha chain EC 2 7 10 1 SEMA domain Description a receptor recognition and occurs in a large family of secreted and binding module found near the transmembrane proteins some of which N terminus of the eukaryotic function
5. in the secretory pathway since the SP value is the highest thus indicating that the sequence contains a signal peptide 2 18 Give both the localisation and the actual signal sequence of AMPL1_SOLLC Leucine aminopeptidase 1 chloroplastic gt sp 010712 AMPL1_SOLLC Leucine aminopeptidase 1 chloroplastic OS Solanum lycopersicum GN LAPA1 PE 2 SV 1 MATLRVSSLFASSSSSLHSNPSVFTKYQSSPKWAFSFPVTPLCSKRSKRIVHCIAGDTLG TRPNESDAPKISIGAKDTAVVQOWQGDLLAIGATENDMARDENSKFKNPLLOQLDSELNG LSAASSEEDFSGKSGQSVNLRFPGGRITLVGLGSSASSPTSYHSLGQAAAAAAKSSQAR NIAVALASTDGLSAESKINSASAIATGVVLGSFEDNRFRSESKKSTLESLDILGLGTGPE IERKIKYAEHVCAGVILGRELVNAPANIVTPAVLAEEAKKIASTYSDVISVNILDAEQC ELKMGAYLAVAAAATENPPYFIHLCFKTPTKERKTKLALVGKGLTFDSGGYNLKVGARS ELMKNDMGGAAAVLGAAKALGEIRPSRVEVHFIVAACENMISAEGMRPGDIVTASNGKI EVNNTDAEGRLTLADALTYACNQGVEKI IDLATLTGAIMVALGP SVAGAF TPNDDLARE FAABEASGEKLWRMPMEES YWESMKSGVADMINTGPGNGGAITGALFLKQFVDEKVOWL DVAGP VWSDEKKNATGYGVSTLVEWVLRN Dm p H H S T SignalP NN prediction Ceuk networks sp Q10712 AMPL1 SOLLC Score a a rf 7 be panne a ar OY ra a maa a O MATLRYSSLFASSSSSLHSNPSVFTKYQSSPKWAFSFPYTPLCSKRSKRI YHCIAGDTLGLTRPNE SDAP Position gt sp_Q10712_AMPL1_ SOLL length 70 Measure Position Value Cutoff signal
6. no strong signal sequence from this entry but whatever signal is present exists at positions 1 19 Name Len mTP SP other Loc RC sp_P16116_ALDR_BOVIN 315 0 085 0 120 m 2 cutoff 0 000 0 000 0 000 The protein is present in the cytoplasm any other location than mitochondria chloroplast and secretory pathway cytoplasm 2 22 23 The major human copper uptake protein hCTR1 is a transmembrane protein 015431 which mediates copper uptake through the cell membrane gt sp 015431 COPT1_HUMAN High affinity copper uptake protein 1 OS Homo sapiens GN SLC31A1l PE 1 SV 1 MDHSHHMGMSYMDSNSTMOQPSHHHPTTSASHSHGGGDSSMMMMPMTF YF GFKNVELLFSG LVINTAGEMAGAF VAVF LLAMF YEGLKIARESLLRKSQVS IRYNSMPVPGPNGTILMETH KTVGQOMLSFPHLLOTVLHIIQVVISYFLMLIFMTYNGYLCIAVAAGAGTGYFLFSWKKA VVVDITEHCH 2 22 Are there any more likely N glycosylation sites except Asn 15 There is an N glycolsolated site at residue112 According to pfam the first transmembrane helix is at residue 60 This means that the first intercellular loop is from res 63 69 the next loop res 86 130 including res 112 would be extracellular again Therefore this site is not to be disregarded NetNGlyc 1 0 predicted N glycosylation sites in sp 015431 COPT1 HUMAN Potential Threshold N glycosylation potential z 20 40 60 86 166 126 146 166 180 Sequence position Threshold 0 5 SeqName Position Potential Jury N Glyc agreem
7. peptide max C 20 0 159 0 32 NO max Y 17 0 205 0 33 NO max S 4 0 632 0 87 NO D 1 16 0 293 0 43 NO The signal sequence is between positions 1 17 However there is no cutoff either at position 17 or 20 since these value don t reach 50 Name Len CTP mTP SP other Loc RC sp_Q10712_AMPL1_SOLL 571 BBM 0 132 0 024 0 032 C 2 cutoff 0 000 0 000 0 000 0 000 The sequence is localized in the chloroplast and is a chloroplast transit peptide highest cTP value 2 19 Give both the localisation and the actual signal sequence of ALDR_BOVIN Aldose reductase gt sp P16116 ALDR_BOVIN Aldose reductase OS Bos taurus GN AKR1B1 PE 1 SV 1 AHNIVLYTGAKMP ILGLGTWKSPPGKVTEAVKVAIDLGYRHIDCAHVYQOQNENEVGLALOA KLOQEQVVKREDLF IVSKLWCT YHDKDLVKGACOKTLSDLKLDYLDLYLIHWPTGEKPGKD FFPLDEDGNVIPSEKDFVDTWTAMEELVDEGLVKAIGVSNEFNHLOVEKILNKPGLKYKPA VNOTECHPYLTQEKLIQYCNSKGIVVTAYSPLGSPDRPWAKPEDPSILEDPRIKAIADKY NKTTAQVLIRFPIQRNLIVIPKSVTPERIAENFOVFDFELDKEDMNTLLSYNRDWRACAL VSCASHRDYPFHEEF SignalP NN prediction Ceuk networks sp P16116 ALDR BOVIN Score AHNIVYLYTGAKMPILGLGTWKSPPGKYTEAYKVAIDLGYRHI DCAHYYQNENEYGLAL QAKLQEQY YKRE Position gt sp_P16116_ALDR_BOVIN length 70 Measure Position Value Cutoff signal peptide max C 26 0 065 0 32 NO max S 4 0 533 0 87 NO mean S 1 18 0 261 0 48 NO D 1 18 0 178 0 43 NO There is
8. 2 15 Which are the advantages and disadvantages of using such sequence pattern databases compared to using databases of amino acid sequences and BLAST A BLAST is the simpler and more straightforward approach since it searches and identifies only the sequences such as amino acids or nucleotides in DNA It is a basic tool PROSITE and similar sequence pattern databases are however more sensitive and generates more data and we are given more information about the protein in itself They give us information on domains families and functional sites as well as the next level of amino acid sequences their patterns signatures and profiles within the protein However the usefulness of each database depends on what you want to study BLAST might be sufficient 2 16 If patterns profiles and HMMs are essentially the same thing made from the same kind of data used for the same purpose etc how come the older methods are still in use Or to put it another way what are the three methods advantages disadvantages when compared to each other A Patterns and profiles have some intrinsic differences While patterns describe some short highly conserved subregions of a protein e g the functional catalytic center of a protein profiles are used to detect domains which are usually far longer and more variable These domains could not be detected with a normal pattern engine but only with more sensitive and robust profiles which allow a certain degree of de
9. Assignment 2 Sequence based predictions This assignment is heavily based on a previous one made by Bengt Persson LiU Introduction This assignment assumes some familiarity with amino acids and their one letter codes Domains are the building blocks of proteins They are regions that fold independently and are often interconnected by flexible linker regions In general each domain is associated with a distinct function for example hydrophobic membrane spanning domains cofactor binding domains and catalytic domains To study relationships between proteins a common first step is to compare the amino acid sequences using a multiple sequence alignment MSA A small example is shown here PORK ORR RRR RRR RRR RK as RRR From the MSA it is possible to determine the conserved regions in the proteins Proteins often contain clusters of residues that are conserved because of particular requirements on their interactions either internally in the protein or with the environment Usually these residues are of functional importance to the protein for example for the binding properties or the catalytic activity The function and structure of proteins can thus be characterized by their conserved sequence motifs and this can be automated using various computational methods e g machine learning techniques In this assignment we will first get acquainted with patterns which is the most basic method for sequence motif recognition Then we w
10. ent result sp_015431_COPT1_HUMAN 15 NSTM 0 7082 9 9 sp_015431_COPT1_HUMAN M2 NGTI 0 5151 5 9 2 23 Does mucin type glycosylation of hCTR1 seem likely No it doesn t seem so likely since the threshold is very low but it might occur and if it does it would occur at positions 17 26 and 27 NetOGlyc 3 1 predicted O glycosylation sites in Sequence Potential Threshold O glycosylation potential 20 46 60 30 100 120 140 166 186 Sequence position Name S T Pos G score I score Y N Comment Sequence T 17 0 532 0 380 T Sequence T 26 0 598 0 069 T Name S T Pos G score I score Y N sp_O15431_C T 17 0 532 0 380 T sp_O15431_C T 26 0 598 0 069 T sp_O15431_C T 27 0 606 0 034 T A recent publication showed that Asn 15 is the only N glycosylation site in hCTR1 and that the key site of mucin type O glycosylation is Thr 27 although additional sites cannot be excluded Both sites where shown to be important for the copper transport function of the protein Links Pfam http pfam sanger ac uk search http pfam sanger ac uk help help Prosite http www expasy ch prosite search http www expasy ch prosite prosuser html user manual SignalP and TargetP http www cbs dtu dk services
11. etween regions of high homology The best alignments of profile and query sequence are calculated according to an overall score The corresponding motif is most likely present in the query sequence when the score exceeds the defined cutoff score 2 10 How do you determine whether or not a sequence matches a profile Are there any differences compared to the pattern case and if so how can they be used to improve the reliability of the results A There are different levels of cut off At lower cut off gives a score of 215 and the results over 215 can have false negative data The higher cut off of 323 and the results above this is the false positive data With this we can compare which sequences are similar Insertions and deletions in the sequence cannot be recognized by the pattern 2 11 Profiles are made from MSAs Give a plausible explanation to how this is done briefly and conceptually max 100 words A Profiles are made from MSA by aligning the amino acid at each position and identifying the position Then a score is given to each amino acid at the particular position A higher score indicates a better match In a deletion state the match state can be omitted and it receives a position dependant penalty Insertion states can also exist and it also gets a postion dependant penalty Hidden Markov models HMM HMMs are in fact very similar to profiles Both have match insert and delete states and both have specific ranking of
12. ill expand our view to the more advanced profiles after which we will move on to the powerful statistical hidden Markov models HMM which represent one of the most sensitive classification methods that exist today Finally we will look at some tools for predicting other protein features Patterns A pattern is usually a number of consecutive residues important for a specific biological function These regions include binding sites or enzymatic catalytic sites Here is one example AG x C x 4 DE This pattern is translated as Ala or Gly any Cys any any any any anything but Glu or Asp Prosite is a well used resource that contains a database of patterns and profiles and also provides web based tools that allows users to analyze proteins online Use the tools in Prosite to scan the protein RON_HUMAN 2 1 Does your protein have any known domains If so which A Two domains are found a SEMA domain and a protein kinase domain 2 2 Which pattern matches the ATP binding region Give both the ID and the actual pattern Hint To see the details for a pattern click the link next to the pattern ID on the form PS00000 A PS00107 PROTEIN_KINASE_ATP _ Protein kinases ATP binding region signature Consensus Pattern LIV G P G P FYWMGSTNH SGA PW LIVCAT PD x GSTACLIVMFY x 5 18 LIVMFYWCSTAR AIVP LIVMFAGCKR K K binds ATP 2 3 Describe how sequences can be matched to this pattern Hint On the docu
13. mentation page you have the option to Retrieve an alignment of Swiss Prot true positive hits which may help The Prosite user manual may also help on this and subsequent questions A Using the consensus sequence we can match sequences for this pattern Leu or Ile or Val Gly anything but Pro Gly anything but Pro Phe or Tyr or Trp or Met or Gly or Ser or Thr or Asn or His Ser or Gly or Ala anything but Pro or Trp Leu or Ile or Val or Cys or Ala or Thr anything but Pro or Asp any Gly or Ser or Thr or Ala or Cys or Leu or Ile or Val or Met or Phe or Tyr any amino acid from five to eighteen positions Leu or Ile or Val or Met or Phe or Tyr or Trp or Cys or Ser or Thr or Ala or Arg Ala or Ile or Val or Pro Leu or Ile or Val or Met or Phe or Ala or Gly or Cys or Lys or Arg Lys 2 4 How do you determine whether or not a sequence matches a pattern A Using a database like PROSITE we can get all the possible sequences for a particular pattern Then we can compare sequences with the pattern for different amino acids in the same location All amino acids have to meet the restrictions of the pattern 2 5 Patterns are made from MSAs Give a plausible explanation as to how this is done A If we have two sequences one with known function and one unknown and these show similarities limited to only a few residues we might need to create a pattern which will give us a model for our que
14. position specific scoring matrices PSSM and as the name suggests they consist of matrices of numbers which are used to score sequences based on which amino acid residue they have where Now getting Prosite to show details on profiles is a bit trickier Go back to the Prosite scan summary and bring up the documentation page for the first matching domain On this page once again click on the link next to the profile ID should have the exact same text as the one you just clicked but will not bring you to the same page If everything went OK you should now be looking at a page with lots and lots of numbers on it 2 9 Given that M means match and I means insertion describe how sequences are matched to this profile A The rows correspond to the AA positions of the profile to which a query sequence should be aligned The columns refer to the AA one letter code including the ambiguous letters B and Z Both thus form a matrix showing position specific substititution scores Furthermore penalties for matchinsertion transitions MI further insertions I and match deletion transitions MD further deletions D and internal initiation termination are defined by default M defines matches in the profile sequence I defines insertions in the profile sequence Alignment of the query sequence with M position is scored specifically according to the position specific substitution values I position oftentimes correspond to gap region b
15. ry sequences This is done by aligning multiple sequences within a protein family which we know are conserved Alignment and subsequent matching will create a model that in turn can be used to find more members of this family containing conserved sequences The pattern is then based on the amino acids within these members However this may exclude atypical proteins which might be homologous while containing modifications acquired during evolution which the other protein members didn t acquire 2 6 Which position in the pattern binds ATP Answer using a number A Position 1114 Lysine K in the protein and the last position in the pattern K depending on how many repeats there are at position 13 binds ATP If we have 5 repeats at position 13 the position in the pattern where ATP binds would be at 21 Consider the following two hypothetical patterns that describe the same conserved region FW C x 3 C AG E MLI D and FW C x 3 C ASGPT E IVLM D 2 7 Which of the patterns is more tolerant A The second one FW C x 3 C ASGPT E IVLM D 2 8 What are the consequences of using a more tolerant pattern in searches A A more tolerant pattern will give more hits as well as a higher sensitivity but lower selectivity Profiles Profiles are more sensitive and more robust than patterns and can therefore be used to describe larger sequence features such as for example domains Profiles are also called
16. viation Modern profile engines e g PROSITE increasing begin to employ HMM like algorithms which are in contrast to classic profiles not only based on experimental results but have a very strong theoretical nathematical fundament are more noise resistant and more reliable as well as flexible than patterns and profiles Other tools There are numerous predictors of protein features available online We only have time to try a few examples A good site with multiple machine learning based predictors is Center for Biological Sequence Analysis in Copenhagen CBS Signal sequences that direct newly synthesised protein to its final localisation where it should fulfill its function are often encoded in the N terminal parts There are now several predictors available for such predictions and SignalP from CBS is the most well known First you shall use CBS predictors to predict the presence or absence of signal sequences and organelle localisation for three sequences below with given Swissprot IDs Hint You first need to get the amino acid sequence Go to UniProt Search for the accession number in question and click FASTA in the upper right corner orange button Use SignalP to predict the signal sequences and TargetP to predict the organelle localisation for the following sequences Give both the localisation and the actual signal sequence 2 17 Give both the localisation and the actual signal sequence of USHA_ECOLI Protein UshA
Download Pdf Manuals
Related Search