Title:SeqMapPDB: A Standalone Pipeline to Identify Representative Structures of Protein Sequences and Mapping Residue Indices in Real-Time at Proteome Scale

Abstract:Motivation: 3D structures of proteins provide rich information for understanding their biochemical roles. Identifying the representative protein structures for protein sequences is essential for analysis of proteins at proteome scale. However, there are technical difficulties in identifying the representative structure of a given protein sequence and providing accurate mapping of residue indices. Existing databases of mapping between structures and sequences are usually static that are not suitable for studying proteomes with frequent gene model revisions. They often do not provide reliable and consistent representative structures that maximizes sequence coverage. Furthermore, proteins isomers are usually not properly resolved. Results: To overcome these difficulties, we have developed a computational pipeline called SeqMapPDB to provide high-quality representative PDB structures of given sequences. It provides mapping to structures that fully cover the sequences when available, or to the set of partial non-overlapping structural domains that maximally cover the query sequence. The residue indices are accurate mapped and isomeric proteins are resolved. SeqMapPDB is efficient and can rapidly carry out proteome-wide mapping to the selected version of reference genomes in real-time. Furthermore, SeqMapPDB provides the flexibility of a stand-alone pipeline for large scale mapping of in-house sequence and structure data. Availability: Our method is available at this https URL with GNU GPL license.