Protein Sequence Analysis — Overview NIH Proteomics Workshop

Protein Sequence Analysis - Overview NIH Proteomics Workshop 2007 Raja Mazumder Scientific Coordinator, PIR Research Assistant Professor, Department of Biochemistry and Molecular Biology Georgetown University Medical Center

Topics l l l Proteomics and protein bioinformatics (protein sequence analysis) Why do protein sequence analysis? Searching sequence databases Post-processing search results Detecting remote homologs

Clinical proteomics From Petricoin et al. , Nature Reviews Drug Discovery (2002) 1, 683 -695

Single protein and shotgun analysis Mixture of proteins Gel based seperation Single protein analysis Shotgun analysis Digestion of protein mixture Spot excision and digestion Peptides from many proteins Peptides from a single protein LC or LC/LC separation MS analysis MS/MS analysis Protein Bioinformatics Adapted from: Mc. Donald et al. (2002). Disease Markers 18: 99 -105

Protein bioinformatics: protein sequence analysis l Helps characterize protein sequences in silico and allows prediction of protein structure and function l Statistically significant BLAST hits usually signifies sequence homology l Homologous sequences may or may not have the same function but would always (very few exceptions) have the same structural fold l Protein sequence analysis allows protein classification

Development of protein sequence databases l Atlas of protein sequence and structure – Dayhoff (1966) first sequence database (pre-bioinformatics). Currently known as Protein Information Resource (PIR) l Protein data bank (PDB) – structural database (1972) remains most widely used database of structures l Uni. Prot – The Universal Protein Resource (2003) is a central database of protein sequence and function created by joining the forces of the Swiss-Prot, Tr. EMBL and PIR protein database activities

Comparative protein sequence analysis and evolution l Patterns of conservation in sequences allows us to determine which residues are under selective constraint (and thus likely important for protein function) l Comparative analysis of proteins is more sensitive than comparing DNA l Homologous proteins have a common ancestor l Different proteins evolve at different rates l Protein classification systems based on evolution: PIRSF and COG

PIRSF and large-scale annotation of proteins l PIRSF is a protein classification system based on the evolutionary relationships of whole proteins l As part of the Uni. Prot project, PIR has developed this classification strategy to assist in the propagation and standardization of protein annotation

Comparing proteins l Amino acid sequence of protein generated from proteomics experiment e. g. protein fragment DTIKDLLPNVCAFPMEKGPCQTYMTRWFFNFETGECELFAYGGCGGNSNNFLRKEKCEKFCKFT l Amino-acids of two sequences can be aligned and we can easily count the number of identical residues (or use an index of similarity) as a measure of relatedness. l Protein structures can be compared by superimposition

Protein sequence alignment l Pairwise alignment a b a c d a b _ c d l Multiple sequence alignment provides more information a b a c d a b _ c d x b a c e l MSA difficult to do for distantly related proteins

Protein sequence analysis overview l Protein databases l l Searching databases l l PIR (pir. georgetown. edu) and Uni. Prot (www. uniprot. org) Peptide search, BLAST search, Text search Information retrieval and analysis l l Protein records at Uni. Prot and PIR Multiple sequence alignment Secondary structure prediction Homology modeling

Universal Protein Resource http: //www. uniprot. org/ Uni. Ref 50 Clustering at 100, 90, 50% Uni. Ref 90 Uni. Prot NREF Uni. Ref 100 Automated Annotation Automated merging of sequences Swiss. Prot Literature-Based Annotation Uni. Prot Knowledgebase Uni. Prot. KB Uni. Prot Archive Uni. Parc Tr. EMBL PIR-PSD Ref. Seq Gen. Bank/ Ens. EMBL/DDBJ PDB Patent Data Other Data

Peptide Search

ID mapping

Query Sequence l Unknown sequence is Q 9 I 7 I 7 l BLAST Q 9 I 7 I 7 against the Uni. Prot Knowledgebase (http: //www. uniprot. org/search/blast. shtml) l Analyze results

BLAST results

Any Field not specific Text search

Text search results: display options specific Move Pubmed ID, Pfam ID and PDB ID into “Columns in Display”

Text search results: add input box

Text search result with null/not null

Uni. Prot beta site http: //beta. uniprot. org/

Uni. Prot. KB protein record

SIR 2_HUMAN protein record

Are Q 9 I 7 I 7 and SIR 2_HUMAN homologs? l Check BLAST results l Check pairwise alignment

Protein structure prediction l Programs can predict secondary structure information with 70% accuracy l Homology modeling - prediction of ‘target’ structure from closely related ‘template’ structure

Secondary structure prediction http: //bioinf. cs. ucl. ac. uk/psipred/

Secondary structure prediction results

Sir 2 structure

Homology modeling http: //www. expasy. org/swissmod/SWISS-MODEL. html

Homology model of Q 9 I 7 I 7 Blue - excellent Green - so so Red - not good Yellow - beta sheet Red - alpha helix Grey - loop

Sequence features: SIR 2_HUMAN

Multiple sequence alignment

Multiple sequence alignment Q 9 I 7 I 7, Q 82 QG 9, SIR 2_HUMAN

Sequence features: CRAA_RABIT

Identifying Remote Homologs

Structure guided sequence alignment