Multiple Sequence Alignment 2.11 2003/01/30 2007/09/12 12:08:35.151 GMT-5 Susan Cates mscates@bioc.rice.edu Susan Cates mscates@bioc.rice.edu bioinformatics clustalw multiple sequence alignment protein sequence sequence comparison This module introduces the reader to multiple sequence alignments. The ClustalW alignment tool is used to align sequences of troponin I from different human tissues and sequences of the cystic fibrosis transmembrane receptor (CFTR) from several different species. Sequence alignments can be used to study the relationship(s) between sequences in sets of more than two sequences. This application is particularly useful when studying the relationships between a similar type of gene product that is expressed by different organisms, like analyzing CFTR sequences from several different species, or when studying similar, yet divergent, sequences within the same organism, as the variance in troponin I isoforms in Homo sapiens. Often, a primary focus of a multiple sequence alignment is to identify, within several related sequences, regions that are highly conserved in identity or similarity, and therefore probably have functional and/or structural significance. Many factors affect the analysis of conserved regions within related sequences, such as the number of sequences included in the analysis, and the ratio of the number of very similar (almost identical) sequences to the number of more distantly related sequences. Divergent sequences can cause problems in a multiple sequence alignment. It is more difficult to identify the correct alignment when two sequences that are related throughout part of the sequence also contain large sections that diverge. Therefore, the error rates in the alignment increase as divergence increases. These errors in the alignment can cause the related part of the sequences to show lower similarity than they actually have, and this sort of error is often amplified in subsequent steps. ClustalW (1), a commonly used multiple sequence alignment program, addresses the problems associated with alignment of divergent sequences in several ways. Individual weights are assigned to each sequence in a partial alignment such that near-duplicate sequences are down-weighted and divergent sequences are up-weighted. Also, the amino acid substitution matrices at different alignment stages are chosen according to the divergence of the sequences to be aligned. Residue-specific gap penalties and locally reduced gap penalties in hydrophilic regions increase the penalty for opening new gaps in regions of regular secondary structure. Therefore, it increases the likelihood that gaps will occur in loop regions than in highly structured regions such as alpha helices and beta sheets. Highly structured regions are commonly very important to the fold and function of a protein, and so divergence is often biologically less likely in these areas. For similar reasons, existing gaps receive locally reduced gap penalties to encourage the opening up of new gaps near the existing ones. These features have been designed into ClustalW to produce multiple sequence alignments that are biologically meaningful. First, use the example of analyzing the variance in troponin I isoforms in Homo sapiens to get acquainted with multiple sequence alignments with ClustalW. ClustalW is available at many bioinformatics web sites, but the EBI site is chosen here, for its nice interface and graphical display. Accept the default values on the submission form, and scroll down the page until the box for pasting in the sequences to be aligned appears. Now, the troponin I sequences must be obtained. For the sake of time, instead of searching with the string "troponin I human", the pertinent accession codes are supplied for this exercise. Open a new browser window to the NCBI home page. Locate the "Search" box at the top of the page and select "Protein" to search the protein database. Type in the accession code TNNI3_HUMAN in the box to the right, and click on "Go". The results of this search will become part of the search history. Select the history link from the menu under the query box to verify this. Now, search "Protein" again, using the accession code TNNI2_HUMAN, then perform one last search using the accession code TNNI1_HUMAN. As an illustration of the search history function in Entrez, combine the results of these three searches. Do this using the boolean operator "OR" in between the number assigned to each search in the history listing. For instance, if the Search History gave the following list, #3 Search TNNI1_HUMAN 17:44:55 #2 Search TNNI2_HUMAN 17:44:49 #1 Search TNNI3_HUMAN 17:41:47 then "#1 OR #2 OR #3" would be the search string. The boolean operator OR is specifying that all results included in any of the above searches be combined. Using the boolean AND operator would specify only results that are common to all three searches, which in this case would yield zero results. To view the results, click on "GO" once you have entered the search string into the query box. Now, the combined search should return all three of the desired troponin I isoforms. Check the box to the left of each accession code for all three results. Just above the list of results, find the region on the left that states "Display Summary" and change "Summary" to "FASTA". This should yield the amino acid sequences in FASTA format for the 3 proteins that were selected. Copy and paste each sequence into the ClustalW window, pressing return after pasting each entry. There does not have to be a line space between entries, as long as the FASTA identifying line starts on a new line. However, it does not hurt to have a line space between entries, either. Note that the numbered search description line from Entrez is not part of the FASTA format. The numbered search description line will look something like this, 1: P19237. Reports Troponin I, slow ...[gi:1351298] and should be omitted from the ClustalW search query. Also, check that the description header for the sequence in FASTA format does not take up more than one line. If it does, shorten the description so that it only occupies one line. Press the "Run" button, and click on the link for the browser window that displays the results, if this option is given. A text page will appear with the alignment scores. What is the score for the alignment of TNNI3 to TNNI2? What is the score for the alignment of TNNI3 to TNNI1? What is the score for the alignment of TNNI2 to TNNI1? At the top of the page, a table contains several choices. There a link to the home page for the alignment editor Jalview,there is a button, labeled "Start Jalview", which is a link to the multiple sequence alignment results in the graphical format produced by the Jalview editor. Choose the "Start Jalview" button to view the alignment results. The sequences are labeled at the left and a bar graph below the sequences indicates which regions have high conservation or similarity. The cardiac isoform of troponin I is roughly 30 amino acids longer than the other two isoforms. Looking at the aligned sequences, in what region are the extra amino acids located in cardiac troponin I? Take a look at the color scheme of the alignment. Regions of the sequence with identical amino acids are colored the same, but notice that regions of the sequence with similar amino acids have also been colored alike. Usually these color schemes have groups consisting of hydrophobic amino acids, hydrophilic amino acids, positively-charged amino acids, and negatively-charged amino acids. Sometimes these large color groups are divided further to distinguish between sidechain lengths, the aromatic sidechains, glycines and prolines. Try one more multiple sequence alignment, analyzing CFTR sequences from several different species. Return to the ClustalW submission page. Cut and paste the CFTR sequences listed below from three species into the query submission box and run the multiple alignment. >gi|203423|gb|AAA40918.1| cftr [Rattus norvegicus] IKHSGRVSFSSQISWIMPGTIKENIIFGVSYDEYRYKSVVKACQLQEDITKFAEQDNTVLGEGGVTLSGG QRARISLARAVYKDADLYLLDSPFGYLDVLTEEQIFESCVCKLMASKTRILVTSKMEQLKKADKILILHE GSSYFYGTFSELQSLRPDFSSKLMGYDTFDQFTEERRSSILTETLRRFSVDDASTTWNKAKQSFRQTGEF GEKRKNSILSSFSSVKKISIVQKTPLSIEGESDDLQERRLSLVPDSEHGEAALPRSNMITAGPTFPGRRR QSVLDLMTFTPSSVSSSLQRTRASIRKISLAPRISLKEEDIYSRRLSQDSTLNITEEINEEDLKECFFDD MVKIPTVTTWNTYLRYFTLHRGLFAVLIWCVLVFLVEVAASLFVLWLLKNNPVNGGNNGTKIANTSYVVV ITSSSFYYIFYIYVGVADTLLALSLFRGLPLVHTLITASKILHRKMLHSILHAPMSTFNKLKAGGILNRF SKDIAILDDFLPLTILT >gi|192832|gb|AAA18903.1| cftr[Mus musculus] MQKSPLEKASFISKLFFSWTTPILRKGYRHHLELSDIYQAPSADSADHLSEKLEREWDREQASKKNPQLI HALRRCFFWRFLFYGILLYLGEVTKAVQPVLLGRIIASYDPENKVERSIAIYLGIGLCLLFIVRTLLLHP AIFGLHRIGMQMRTAMFSLIYKKTLKLSSRVLDKISIGQLVSLLSNNLNKFDEGLALAHFIWIAPLQVTL LMGLLWDLLQFSAFCGLGLLIILVIFQAILGKMMVKYRDQRAAKINERLVITSEIIDNIYSVKAYCWESA MEKMIENLREVELKMTRKAAYMRFFTSSAFFFSGFFVVFLSVLPYTVINGIVLRKIFTTISFCIVLRMSV TRQFPTAVQIWYDSFGMIRKIQDFLQKQEYKVLEYNLMTTGIIMENVTAFWEEGFGELLEKVQQSNGDRK HSSDENNVSFSHLCLVGNPVLKNINLNIEKGEMLAITGSTGSGKTSLLMLILGELEASEGIIKHSGRVSF CSQFSWIMPGTIKENIIFGVSYDEYRYKSVVKACQLQQDITKFAEQDNTVLGEGGVTLSGGQRARISLAR AVYKDADLYLLDSPFGYLDVFTEEQVFESCVCKLMANKTRILVTSKMEHLRKADKILILHQGSSYFYGTF SELQSLRPDFSSKLMGYDTFDQFTEERRSSILTETLRRFSVDDSSAPWSKPKQSFRQTGEVGEKRKNSIL NSFSSVRKISIVQKTPLCIDGESDDLQEKRLSLVPDSEQGEAALPRSNMIATGPTFPGRRRQSVLDLMTF TPNSGSSNLQRTRTSIRKISLVPQISLNEVDVYSRRLSQDSTLNITEEINEEDLKECFLDDVIKIPPVTT WNTYLRYFTLHKGLLLVLIWCVLVFLVEVAASLFVLWLLKNNPVNSGNNGTKISNSSYVVIITSTSFYYI FYIYVGVADTLLALSLFRGLPLVHTLITASKILHRKMLHSILHAPMSTISKLKAGGILNRFSKDIAILDD FLPLTIFDFIQLVFIVIGAIIVVSALQPYIFLATVPGLVVFILLRAYFLHTAQQLKQLESEGRSPIFTHL VTSLKGLWTLRAFRRQTYFETLFHKALNLHTANWFMYLATLRWFQMRIDMIFVLFFIVVTFISILTTGEG EGTAGIILTLAMNIMSTLQWAVNSSIDTDSLMRSVSRVFKFIDIQTEESMYTQIIKELPREGSSDVLVIK NEHVKKSDIWPSGGEMVVKDLTVKYMDDGNAVLENISFSISPGQRVGLLGRTGSGKSTLLSAFLRMLNIK GDIEIDGVSWNSVTLQEWRKAFGVITQKVFIFSGTFRQNLDPNGKWKDEEIWKVADEVGLKSVIEQFPGQ LNFTLVDGGYVLSHGHKQLMCLARSVLSKAKIILLDEPSAHLDPITYQVIRRVLKQAFAGCTVILCEHRI EAMLDCQRFLVIEESNVWQYDSLQALLSEKSIFQQAISSSEKMRFFQGRHSSKHKPRTQITALKEETEEE VQETRL >gi|6862589|gb|AAF30300.1|AAF30300 cftr [Mus musculus] MQKSPLEKASFISKLFFSWSTAILRKGYRQHLELSDIYQAPSADSADHLSEKLEREWDREQASKKNPQLI HALRRCFFWRFLFYGILLYLGEVTKAVQPVLLGRIIASYDPENKVERSIAIYLGIGLCLLFIVRTLLLHP AIFGLHRIGMQMRTAMFSLIYKKTLKLSSRVLDKISIGQLVSLLSNNLNKFDEGLALAHFIWIAPLQVTL LMGLLWDLLQFSAFCGLGLLIILVIFQAILGKMMVKYRDQRAAKINERLVITSEIIDNIYSVKAYCWESA MEKMIENLREVELKMTRKAAYMRFFTSSAFFFSGFFVVFLSVLPYTVINGIVLRKIFTTISFCIVLRMSV TRQFPTAVQIWYDSFGMIRKIQDFLQKQEYKVLEYNLMTTGIIMENVTAFWEEGFGELLEKVQQSNGDRK HSSDENNVSFSHLCLVGNPVLKNINLNIEKGEMLAITGSTGSGKTSLLMLILGELEASEGIIKHSGRVSF CSQFSWIMPGTIKENIIFGVSYDEYRYKSVVKACQLQQDITKFAEQDNTVLGEGGVTLSGGQRARISLAR AVYKDADLYLLDSPFGYLDVFTEEQVFESCVCKLMANKTRILVTSKMEHLRKADKILILHQGSSYFYGTF SELQSLRPDFSSKLMGYDTFDQFTEERRSSILTETLRRFSVDDSSAPWSKPKQSFRQTGEVGEKRKNSIL NSFSSVRKISIVQKTPLCIDGESDDLQEKRLSLVPDSEQGEAALPRSNMIATGPTFPGRRRQSVLDLMTF TPNSGSSNLQRTRTSIRKISLVPQISLNEVDVYSRRLSQDSTLNITEEINEEDLKECFLDDVIKIPPVTT WNTYLRYFTLHKGLLLVLIWCVLVFLVEVAASLFVLWLLKNNPVNSGNNGTKISNSSYVVIITSTSFYYI FYIYVGVADTLLALSLFRGLPLVHTLITASKILHRKMLHSILHAPMSTISKLKAGGILNRFSKDIAILDD FLPLTIFDFIQLVFIVIGAIIVVSALQPYIFLATVPGLVVFILLRAYFLHTAQQLKQLESEGRSPIFTHL VTSLKGLWTLRAFRRQTYFETLFHKALNLHTANWFMYLATLRWFQMRIDMIFVLFFIVVTFISILTTGEG EGTAGIILTLAMNIMSTLQWAVNSSIDTDSLMRSVSRVFKFIDIQTEESMYTQIIKELPREGSSDVLVIK NEHVKKSDIWPSGGEMVVKDLTVKYMDDGNAVLENISFSISPGQRVGLLGRTGSGKSTLLSAFLRMLNIK GDIEIDGVSWNSVTLQEWRKAFGVITQKVFIFSGTFRQNLDPNGKWKDEEIWKVADEVGLKSVIEQFPGQ LNFTLVDGGYVLSHGHKQLMCLARSVLSKAKIILLDEPSAHLDPITYQVIRRVLKQAFAGCTVILCEHRI EAMLDCQRFLVIEESNVWQYDSLQALLSEKSIFQQAISSSEKMRFFQGRHSSKHKPRTQITALKEETEEE VQETRL >gi|1669377|gb|AAB46340.1| cftr [Homo sapiens] LLLIVIGAIAVVAVLQPYIFVATVPVIVAFIMLRAYFLQTSQQLKQLESEGRSPIFTHLVTSLKGLWTLR AFGRQPYFETLFHKALNLHTANWFLYLSTLRWFQMRIEMIFVIFFIAVTFISILTTGEGEGRVGIILTLA MNIMSTLQWAVNSSIDVDSLMRSVSRVFKFIDMPTEGKPTKSTKPYKNGQLSKVMIIENSHVKKDDIWPS GGQMTVKDLTAKYTEGGNAILENISFSISPGQRVGLLGRTGSGKSTLLSAFLRLLNTEGEIQIDGVSWDS ITLQQWRKAFGVIPQKVFIFSGTFRKNLDPYEQWSDQEIWKVADEVGLRSVIEQFPGKLDFVLVDGGCVL SHGHKQLMCLARSVLSKAKILLLDEPSAHLDPVTYQIIRRTLKQAFADCTVILCEHRIEAMLECQQFLVI EENKVRQYDSIQKLLNERSLFRQAISPSDRVKLFPHRNSSKCKSKPQIAALKEETEEEVQDTRL >gi|180332|gb|AAA35680.1| cftr [Homo sapiens] MQRSPLEKASVVSKLFFSWTRPILRKGYRQRLELSDIYQIPSVDSADNLSEKLEREWDRELASKKNPKLI NALRRCFFWRFMFYGIFLYLGEVTKAVQPLLLGRIIASYDPDNKEERSIAIYLGIGLCLLFIVRTLLLHP AIFGLHHIGMQMRIAMFSLIYKKTLKLSSRVLDKISIGQLVSLLSNNLNKFDEGLALAHFVWIAPLQVAL LMGLIWELLQASAFCGLGFLIVLALFQAGLGRMMMKYRDQRAGKISERLVITSEMIENIQSVKAYCWEEA MEKMIENLRQTELKLTRKAAYVRYFNSSAFFFSGFFVVFLSVLPYALIKGIILRKIFTTISFCIVLRMAV TRQFPWAVQTWYDSLGAINKIQDFLQKQEYKTLEYNLTTTEVVMENVTAFWEEGFGELFEKAKQNNNNRK TSNGDDSLFFSNFSLLGTPVLKDINFKIERGQLLAVAGSTGAGKTSLLMMIMGELEPSEGKIKHSGRISF CSQFSWIMPGTIKENIIFGVSYDEYRYRSVIKACQLEEDISKFAEKDNIVLGEGGITLSGGQRARISLAR AVYKDADLYLLDSPFGYLDVLTEKEIFESCVCKLMANKTRILVTSKMEHLKKADKILILNEGSSYFYGTF SELQNLQPDFSSKLMGCDSFDQFSAERRNSILTETLHRFSLEGDAPVSWTETKKQSFKQTGEFGEKRKNS ILNPINSIRKFSIVQKTPLQMNGIEEDSDEPLERRLSLVPDSEQGEAILPRISVISTGPTLQARRRQSVL NLMTHSVNQGQNIHRKTTASTRKVSLAPQANLTELDIYSRRLSQETGLEISEEINEEDLKECLFDDMESI PAVTTWNTYLRYITVHKSLIFVLIWCLVIFLAEVAASLVVLWLLGNTPLQDKGNSTHSRNNSYAVIITST SSYYVFYIYVGVADTLLAMGFFRGLPLVHTLITVSKILHHKMLHSVLQAPMSTLNTLKAGGILNRFSKDI AILDDLLPLTIFDFIQLLLIVIGAIAVVAVLQPYIFVATVPVIVAFIMLRAYFLQTSQQLKQLESEGRSP IFTHLVTSLKGLWTLRAFGRQPYFETLFHKALNLHTANWFLYLSTLRWFQMRIEMIFVIFFIAVTFISIL TTGEGEGRVGIILTLAMNIMSTLQWAVNSSIDVDSLMRSVSRVFKFIDMPTEGKPTKSTKPYKNGQLSKV MIIENSHVKKDDIWPSGGQMTVKDLTAKYTEGGNAILENISFSISPGQRVGLLGRTGSGKSTLLSAFLRL LNTEGEIQIDGVSWDSITLQQWRKAFGVIPQKVFIFSGTFRKNLDPYEQWSDQEIWKVADEVGLRSVIEQ FPGKLDFVLVDGGCVLSHGHKQLMCLARSVLSKAKILLLDEPSAHLDPVTYQIIRRTLKQAFADCTVILC EHRIEAMLECQQFLVIEENKVRQYDSIQKLLNERSLFRQAISPSDRVKLFPHRNSSKCKSKPQIAALKEE TEEEVQDTRL >gi|1809238|gb|AAB46352.1| cftr [Homo sapiens] MQRSPLEKASVVSKLFFSWTRPILRKGYRQRLELSDIYQIPSVDSADNLSEKLEREWDRELASKKNPKLI NALRRCFFWRFMFYGIFLYLGEVTKAVQPLLLGRIIASYDPDNKEERSIAIYLGIGLCLLFIVRTLLLHP AIFGLHHIGMQMRIAMFSLIYKKTLKLSSRVLDKISIGQLVSLLSNNLNKFDEGLALAHFVWIAPLQVAL LMGLIWELLQASAFCGLGFLIVLALFQAGLGRMMMKYRDQRAGKISERLVITSEMIENIQSVKAYCWEEA MEKMIENLRQTELKLTRKAAYVRYFNSSAFFFSGFFVVFLSVLPYALIKGIILRKIFTTISFCIVLRMAV TRQFPWAVQTWYDSLGAINKIQDFLQKQEYKTLEYNLTTTEVVMENVTAFWEEGFGELFEKAKQNNNNRK TSNGDDSLFFSNFSLLGTPVLKDINFKIERGQLLAVAGSTGAGKTSLLMVIMGELEPSEGKIKHSGRISF CSQFSWIMPGTIKENIIFGVSYDEYRYRSVIKACQLEEDISKFAEKDNIVLGEGGITLSGGQRARISLAR AVYKDADLYLLDSPFGYLDVLTEKEIFESCVCKLMANKTRILVTSKMEHLKKADKILILHEGSSYFYGTF SELQNLQPDFSSKLMGCDSFDQFSAERRNSILTETLHRFSLEGDAPVSWTETKKQSFKQTGEFGEKRKNS ILNPINSIRKFSIVQKTPLQMNGIEEDSDEPLERRLSLVPDSEQGEAILPRISVISTGPTLQARRRQSVL NLMTHSVNQGQNIHRKTTASTRKVSLAPQANLTELDIYSRRLSQETGLEISEEINEEDLKECFFDDMESI PAVTTWNTYLRYITVHKSLIFVLIWCLVIFLAEVAASLVVLWLLGNTPLQDKGNSTHSRNNSYAVIITST SSYYVFYIYVGVADTLLAMGFFRGLPLVHTLITVSKILHHKMLHSVLQAPMSTLNTLKAGGILNRFSKDI AILDDLLPLTIFDFIQ Before looking at the alignment in Jalview, look at the text results. Which 2 sequences align with the highest pairwise alignment score? Look at the text display of the alignments, shown just below the scores. Try to determine the regions where the 6 sequences align most closely. Many people find this type of text display hard to interpret (a fact that is good to remember when putting together a figure for a presentation or an article). At the bottom of the text display is a cladogram. A cladogram is a branched phylogenetic-type tree where the branches are of equal length. Cladograms can show common ancestry, but because the branches have equal lengths, they do not provide an accurate indication of the evolutionary distance between the branches. Does the cladogram group together all the human CFTR sequences in the same branch? Now, look at the alignment in Jal view. Is it easier to identify the regions where all 6 sequences align most closely in the Jalview display? Leave both of these windows open (closing the ClustalW text results browser page may cause the Jalview editor page to disappear automatically). Notice the file number assigned this Jalview sequence alignment. This is to help distinguish it for purposes of comparison from the next and last alignment in this tutorial, where the gap penalty will be manipulated from the default values. Finally, repeat the CFTR alignment using the sequences above, but this time change "Gap Open" to 100. (Click on the term "Gap Open" to view the default value.) This will increase the penalty for opening a gap. For more information on gaps and gap penalties, view the EBI help page entitled "About Gaps". Compare the group alignment scores from the previous alignment with these group alignment scores. The group alignment scores can be found in the .output file (listed in the results section) under the section which starts: Start of Multiple Alignment There are 5 groups Aligning... Does changing the gap open penalty affect the group alignment scores? Are the group scores higher or lower when the penalty is increased to 100? View the Jalview display of the new alignment. Compare the Jalview displays of the first alignment, and the new alignment with the increased gap penalty, in the regions numbered 1220 to 1250. Describe the differences in the two alignments in this region, both in terms of the gaps, and in terms of the similarity bar graph displayed below the sequences. A look at the ClustalW submission page is enough to see that there are many more parameters that can be manipulated by the user. However, the default values have been optimized and generally should be left alone, except in cases where the user has a definitive justification for making the change. It does frequently occur in biology, though, that the user knows some empirical information that conflicts with the results given by an alignment performed with the default values. In these cases, a scientific argument can be made for altering the parameters to force the alignment to reflect the empirical information. Thompson J.D., Higgins D.G., Gibson T.J. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice Nucleic Acids Res. 1994 22:4673-4680