Premium
Protein Function Prediction Using ProMOL and PyMOL
Author(s) -
Hart Kaitlin,
McKay Talia,
TedlaBoyd Weinishet,
Mills Jeffrey,
Bernstein Herbert,
Craig Paul
Publication year - 2015
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.29.1_supplement.722.5
Subject(s) - protein data bank , protein data bank (rcsb pdb) , computational biology , molecular graphics , computer science , protein structure , levenshtein distance , structural bioinformatics , protein function , sequence alignment , protein structure database , structural alignment , plug in , bioinformatics , data mining , computer graphics , biology , peptide sequence , sequence database , genetics , artificial intelligence , programming language , biochemistry , gene
We use ProMOL and PyMOL to predict the function of structures that lack functional annotation. There are currently over 3700 proteins in the Protein Data Bank that do not have an assigned function. ProMOL, a plugin for the molecular graphics system PyMol, was developed for the purpose of characterizing protein structures of unknown function. The program was created, and is constantly being revised, by members of the RIT and Dowling College communities. ProMOL compares query protein structures against a library of active site motif templates. ProMOL calculates Levenshtein distances and RMSD values, and generates a 3D residue alignment visual between the motif active site residues and corresponding query residues. The protein queries which generate the best results are then submitted to NCBI's BLAST algorithm, the University of Helsinki's Dali alignment tool, and Sanger Institute's Pfam database. If the data produced by these four bioinformatics programs lead to a preliminary conclusion concerning the function of an uncharacterized structure, the behavior of the protein is then tested in vitro . To date, there have been over 90 significant “hits” for over 70 uncharacterized protein structures using ProMOL's motif library. These structures are currently being analyzed via BLAST, Dali, and Pfam. While computational data analysis in an ongoing process, the most promising alignments to date include PDB entries 2HRZ, 4EZI, 3L1W, and 2O14. This work has been supported in part by National Institute of General Medical Sciences grants 2R15GM078077‐02, 3R15GM078077‐02S1, and 3R15GM078077‐02S2.