Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

PAIRWISE ALIGNMENT

Pairwise Sequence Alignment is used to identify regions of similarity that may indicate functional,

structural and/or evolutionary relationships between two biological sequences (protein or nucleic acid).

By contrast, Multiple Sequence Alignment (MSA) is the alignment of three or more biological sequences

of similar length. From the output of MSA applications, homology can be inferred and the evolutionary

relationship between the sequences studied.

Pairwise alignment is performed using an algorithm known as Dynamic Programming (DP), which has 2

main variants:

- Global alignment (Needleman-Wunsch) — the sequences are forced to be aligned across their

entire length. (http://www.ebi.ac.uk/Tools/psa/emboss_needle/help/index-protein.html)

- Local alignment (Smith-Waterman) — only the best matching sub-part of the sequences are

aligned).

(http://www.ebi.ac.uk/Tools/psa/emboss_water/help/index-protein.html)

Activity 1: Write a brief description about the two proteins P29600 and

P41363

P29600:

Is a protein called Subtilisin Savinase from the organism Lederbergia lentus (Bacillus lentus). It’s an

extracellular alkaline serine protea and its main function is to catalyzes the hydrolysis of proteins and

peptide amides.

P41363:

Is a protein called Thermostable alkaline protease that is coded from the gene BH0855 on the organism

Alkalihalobacillus halodurans (strain ATCC BAA-125 / DSM 18197 / FERM 7344 / JCM 9153 / C-125)

(Bacillus halodurans). Its only known function is that it shows keratinolytic activity.

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

Activity 2: Global Alignment

1. Go to EBI Pairwise Sequence Alignment (PSA) at

https://www.ebi.ac.uk/Tools/psa/emboss_needle/,

2. Go to Global Alignment ➔ Needle (EMBOSS) ➔ Protein Needle (Short for Needleman &

Wunsch)

3. Paste in the sequences - including the header line - one sequence in each search

4. box. Set format to pair

5. Go to "STEP 2" ➔"More options" ➔ Matrix and set to "BLOSUM62";

6. Set End Gap Penalty to "true" (otherwise, the algorithm ignores gaps at the very ends of the

alignments, i.e. gaps positioned at the end of one of the sequences would not cost anything).

7. Finally, start the alignment by clicking the button labeled "Submit"

RESULTS OF THE ALIGNMENT:

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

1. What is the Alignment score?

860.5

2. What is the Alignment length?

361

3. What is the % and fraction Identity? (The value reported for "Identity" includes

perfect matches only)

176/269 (65.4%)

4. What is the % and fraction Similarity? (The value reported for "Similarity" includes

perfect matches + "close" mismatches)

214/269 (79.6%)

Results from NCBI:

Activity 3: Local Alignment

1. Go to https://www.ebi.ac.uk/Tools/psa/ ➔ Local Alignment ➔ Water (EMBOSS)

2. Select protein alignment.

3. Use protein set A

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

1. What is the Alignment score?

916.0

2. What is the Alignment length?

269

3. What is the % and fraction Identity? (The value reported for "Identity" includes

perfect matches only)

176/269 (65.4%)

4. What is the % and fraction Similarity? (The value reported for "Similarity" includes

perfect matches + "close" mismatches)

214/269 (79.6%)

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

5. Do you think one of them is more biologically relevant than the other, or do both

contribute valuable information?

Since the two sequences are of different lengths makes the most sense to use the Smith-

Waterman algorithm ("local alignment"), as this will provide an analysis of differences and

similarities for that part of the sequence which is actually comparable.

The sequences are very similar so there generally are not much difference in the information

you get from using local and global alignment.

6. How were the amino acid sequences of the two proteins determined? (Hint: look

at the titles of the papers, and the Cited for fields, listed in the Reference

sections).

The P29600 sequence is derived from 3D structure.

And P41363 is translated from DNA + information from protein sequencing.

7. Subcellular localization: Where in (or outside) the cell do the enzymes function?

Both of the proteins are Secreted proteins.

8. The feature table contains details about the regions/domains of the protein - try

to do a comparison to spot the differences between the two UniProt entries (Hint:

focus on the "PTM / processing" section).

P29600 starts with the sequence of the mature protein.

P41363 starts with a signal peptide (pos: 1-24), then pro-peptide (25-93), and then comes the

mature protein.

Both signal peptide (function: signal for export of the protein) and the pro-peptide (function:

helps protein to fold correctly or make sure the protein is not active before it is in the correct

place - especially important for proteases that should not start to do its function before being

activated.

Activity 4: Exploring Gaps

We have been comparing sequences that are relatively closely related. Let’s

examine what happens when we compare human protein (P07498) with mice protein

(P06796) and, or prokaryotic Savinase (P29600) protease with a human peptidase

(P29144).

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

- With protein sequences sets B and, or C. Perform global alignment (Needle)

- Set End Gap Penalty to "true":

1. What is the Alignment score?

276.5

2. What is the Alignment length?

193

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

3. What is the % and fraction Identity? (The value reported for "Identity" includes

perfect matches only)

Identity: 76/193 (39.4%) and Similarity: 106/193 (54.9%)

4. What is the % and fraction Similarity? (The value reported for "Similarity" includes

perfect matches + "close" mismatches)

23/196 (11.96%)

- Repeat the alignment with End Gap Penalty set to "false" and report the same

results as above.

The results are the same as reported above.

- Repeat the alignment again — this time using the local alignment algorithm

(Water) — and report the same results as above.

The results are the same as reported above

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

Do you think local or global alignment is best for finding similar parts of distantly related

proteins? Why? Hint: Distantly related proteins typically share a core that relates to the

function of the protein

The prokaryotic protease matches only a single area in the middle of the human protease. In contrast,

this is not clear in the global alignment with end gaps, which "spreads" the short sequence over the

entire length. For distantly related sequences, it would be best to use local alignment because it would

actually provide an optimal analysis of the comparable part of the sequences.

Open the "Shuffle Protein" page

(http://www.bioinformatics.org/sms2/shuffle_protein.html) in a new window/tab, paste

in the tripeptidyl peptidase sequence and shuffle it.

Shuffle Protein results:

Version1:

>results for 1249 residue sequence "sp|P29144|TPP2_HUMAN Tripeptidyl-peptidase 2 OS=Homo sapiens OX=9606

GN=TPP2 PE=1 SV=4" starting

"MATAATEEPF"FCLKCEETKLGSQKVMGSTVNQKKKIVNSEKHGPYNVEKIIAVSLNFCNVILSGIARDGRDNHDVLARPLDHSYEKNENEQNPLYPDNKFHGSMNLAAN

AVVIHHSLALRIDQKCGSKPPTLLDAYDVYQVCKGEIGERPPVSLSELESKWPSITELKMPSGYISTGGWRLIGTEEDFEWIQLPTMTCSSTEWNNFRDLEGPKISELLKG

TELSDPWYLYFHRLLDLELLDWKLSEKGFTVIFYETYGKQLSHGKTETTYHAYGKKSLLSVSQLSDRDNRSRVDVIWIALALGAINAIVTTNSVKVAAVKALSWVVKATFG

CIYVTGPDSRKKAIRAYVGDGFRIYDVLTAQAMKNTGLTRMVLFNSAYKAGLSDNYVPNKLGDSAERHGRTDITAELYTEGLRNVSTAGETYLPDAAPGTKLLDAKNDPSG

PDQRAYSYKSLSLNKGYVSARSTPSDGVQANGRKHSLRIISAAAIVSSPPGYIKLPEEFPFYSNHCVQFITTERPNPETQVCWEITCRSTKTLGLMRQSAALLTIANLSFT

VLGQGLDGQCGTTGGLANELQSEDRKPPGILLRSCINTRKGPPEQNLEALWKTTLPVIIEHHGDESKSNFETPFYKTIQHLVEHQHHNKYIVYTTLSHEGSEAIYASRRPK

WNGAGANYELENFRYHEILGDFPQADGATGEDIEKLANVIEGYRIEESSTTQAGLVSKGIGVKAFKDLSITWMPVELPDHNARLDKKLYIASGVEGVITLDICEQGLKICV

LQQDPVPADKPFCYHSEFQPSATLIPFIDDRSEEDNIGGPDIMTVNMIAEQPVAIRQYDSSVHKTAAFFANSATAMPFAGMRVIAHRTPAGHYDLQENASSWVVSCVVWTE

EWDLRGRSSLLTGMLNTNVIVSFKSATFKHGAVSDHTDAITDCRPTVPDSKGACQKGYEEKMIEGLASIELKTHKANADVEDFPQNPPGLYVHKADLCVQNEGLAIDIKVL

HLPNFGRDNREILALPNSSVAKEPLNHHAVCKIISAISSNKDPFLEIKIEGATVVFKKTVSTEGTDKQLVGYFPLFTNPSGADGNNPMVLPLANENLIKDKGYIYDRPSDG

VTVCTAYGALGQKGACWQYIASWAMDKYSGSPDAYNKVYKLPEKVILLEIWPSSVCMRAPEKHDFEVCPEENLEGGMAAQLTKLPVNVLDQTKCYDLHNELHGVEYGFGRL

VIPIEQFSADNDKKAERVWSAQTKAPPLLVMDVCEWHAHL

Version 2:

>results for 1249 residue sequence "sp|P29144|TPP2_HUMAN Tripeptidyl-peptidase 2 OS=Homo sapiens OX=9606 GN=TPP2 PE=1

SV=4" starting "MATAATEEPF"

NHHIAIDKEPRWLVADTKEASNGVAVNVYIGKEIADLTEGAGLDVTAYCADNPELSGPGDSYEIDARPQKEKGSPMALGWAVPKWEDVDVVAPKICTIGAAIYELRINDAVQGFKELCP

PTTKIHTNPIWASDNSIWNLCDKKIGSMDISDFFVDQFFEDEYHYPMAEGLTLPDLMSGDWEQINKYMITKESSIMAPCNRAYLTWLKGTDCSVRQAMPGWDIPHSAVSMSNLCFNVSV

QVTEHSLVSIVKNKRAPAGQLQLKLSSKDTDGARPVGIYRNKAEEHSTLIGEAIVSLLGLQINHKIYSCKIAPTASDEWLPKGYERSREKGGKSYEDRTTINTKVGEISVWSDRADYDI

NFEVQITVVPEAPYPYGTLCATISNSRVVRGVALHDLLPYFEGHKQYITQVKKRDCPDMIGMLKRSTVLSEFSVSQTSIDGYSFDRSGERDMISAYAIEHNFQQGDQLEPYKGLKGNTI

VKVGGLGENQKLPCPLGNIAAQHDTYFVSYNVALLIAVQGPAPDCAPPNLLLADWVVTNIVSNEIPMKTTASLTCNLDFKHDPTGDSASPLGEGESYVVLLIAKDVEKGAYALKGDCAV

NDYSIFTLIYKLQYPYAERYSGTKAIVRSPCYKQPRLKDLATELGKFTANNDNLHHSTDEGPTVRLCPRSFELSKQPHCPSEMGWRSNPWSGLLEVIAKELGTTRTGKIPSYPALLLGD

LPQLLLVEAKLVGSRDLTFSKTFTVSGPVRDGDIEWHIQSEAYKCEKAKREAFAGIPSGRTEPKTFHLLPACSFSAAAGKDIITAKAWGGHLQSPKSTIVWDGYNSTTFPACNHAKEEL

VSPQEINDWISWIGKIRELHDDSLKLYAPMVKTPGTASGTVTNAAGDDNFGDMPTPNTKKLLEVRNMNIEKTLQTSGQCPLAEANGIRLASALLSYQVPLNLLVQVRTILGTSIVEAFG

WHCPLLHEILKKRNLDLTDVINAAITVFQSDDIKSDGAYINTQETPAAKSQATHALSWRADGLAEQHNHNSSKKNTKGSCVVGHRGFPGLACMEFEVETSKKKKVEFTMGVFLYVDEVK

EKLTGLYYDELEAFSSANIARHLLLYGNILILGEALVYEDQSRTVFEQKLKIPTPYSQFPARETHCHKQNRSGRSNKETLNHESYNVHEHVLSDYINVNEEEVDSNHHLEVVIRSGLGG

FYEPGLEKV CYRTPKNNNGALFLPNFRELSGLAEYPLFMHQRIKRVVTFVEGVNAHGK

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

Version 3:

>results for 1249 residue sequence "sp|P29144|TPP2_HUMAN Tripeptidyl-peptidase 2 OS=Homo sapiens OX=9606 GN=TPP2 PE=1

SV=4" starting "MATAATEEPF"

GTFTNDPDPNCIRIIYPSTAKMEGTNKHDVEGALYKEAVYRAVDPFGREGYHTDFTIGPQGQLYSPDYCQGAPTGNALLVNPEIAQVYTVLIPDTIEEKTGDSRLTVFAKVGKSTFKYL

SCINWPKARMAATFWGWLTNLLKVFDVKLNYCEPQVTDAMLPNAIIDECISGSREESRSGSESIIDHNGYNVPSSVELGETQMLKTEEVQLAAEDKAYDRGGEGISPAVIVFEKYGSLE

LSKSNIISGEYVLSFTVRHHSKPGCYYPQAPENWHSLHKNVRVSRASSLLDPALQTQLPLYPTMRVAKTWQSYNIKPEALPEATHDKIVLVKEDFIVKMLKKENAGGNEKYSGRVSLKS

APWAEIEGYATAAFVDKNRAATSEITGHSKIADVRKIQRGHRTYGGFVLPDYALPWPGTPPHVALIASADRGVSDQPKDRHESIDIVLSGIKFRYLQDTIAIAHISLIVELGKLYDVNG

AQPLQGELWCDGLLTALGKVQTANLSEAFKIVHVETYRMTIKVNEIKAEHKVSEPELNWKAPLCNDQIHGYRSSLFKSSAYDRGALLSLCISKGRNSEIKAVEGSEGNAFGTRPNTIVM

AGDGHKESVTSLDMDWLTRYPEDRAIETYVETTHFYLLLEELNIVYAVFGEGALTNDQFNDTKSFMTQAKISVLPDSNNQDKLDVNNVPQILSALAWANCGVNGDLQLLDHQPATLLLL

KLARCVKFQILMKQKALRGQQFKCPKYPYTVTVIWTRTTPYDFAENEGYIIPINAGGPPAMVLGLLNSRDCSPHVWNIVWESVDAVTLISYGHTILLLTVTGGLFVAQHTDGNTGRYAR

LSVPSPNKSWLLDKQGVTYELGAVVEEPQPEKDGEFGCVELSESWAMKESWNNAGLTLSIALSIMGNVVENSPVESKVSDSFPANFMITVTKADPHAKKDQHAHRGVSNHNPRNITGFC

RNPYEESLGTGQPGALAADENLIHTTCYTKHLWSLHRLECDGMVASPVLTNTNLKLIEIKEFKEKGKKFHRPGSPRLSCDCLAPADVDSINAWLHEETKHTLLLFTKFNGRIICAIDIS

LSDKSILCALVDSANDLCDDNYQTNQSSQPILRMLCSIARGKEENDPCASVARRPSYTGKHKVHCAPEGGYSAADKEVGPEAHYGLAEEKHFSSLFVGQKGKFEECSDEFGPDYESRVV

GAFKNGEPPEKTLNKDQDDIVQMSSIMANKTIGIYDHAKILSWYIKGLPRYTTVLPDKK

Then align Savinase and the shuffled tripeptidyl peptidase sequence using local alignment:

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

Repeat the above procedure two more times, so that you align Savinase with three

different shuffled versions of tripeptidyl peptidase.

For version 2:

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

For version 3:

1. How do the local alignments look? (What are the ranges of Alignment score,

Alignment length, Identity, Similarity, and gap percentage)?

We can see the results from each alignment in the screenshots above.

The results are different for every Alignemnt because the shuffled sequences are all different.

If we had completed the experiment 100 times instead of only 3 times, we could have done

statistics on the outcome and calculated confidence limits and therefore evaluated the degree

of statistical significance from a given alignment score (more on statistical significance when we

come to BLAST).

Bioinformatics and Functional Genomics Student ID: 21485094

Lecturer: Dr. Obed Brew Beatriz Manso

2. Comparing the Savinase/shuffled alignment to the previous Savinase/Human

Peptidase alignment - how will you judge the alignment with human peptidase

now? (More/Less confidence in relation between the sequences?).

The score is higher with the Savinase/Human Peptidase alignment than what we got with the

shuffled sequences. However, only the scores show a clear difference between them.