作者: Genki Terashi , Mayuko Takeda-Shitaka
DOI: 10.1371/JOURNAL.PONE.0141440
关键词:
摘要: Proteins are flexible, and this flexibility has an essential functional role. Flexibility can be observed in loop regions, rearrangements between secondary structure elements, conformational changes entire domains. However, most protein alignment methods treat structures as rigid bodies. Thus, these fail to identify the equivalences of residue pairs regions with flexibility. In study, we considered that evolutionary relationship proteins corresponds directly residue–residue physical contacts rather than three-dimensional (3D) coordinates proteins. developed a new method, contact area-based (CAB-align), which uses area similarity. The main purpose CAB-align is homologous relationships at level related structures. procedure comprises two steps: First, rigid-body method based on local global 3D superposition employed generate sufficient number initial alignments. Then, iterative dynamic programming executed find optimal alignment. We evaluated performance advantages four points: (1) agreement gold standard alignment, (2) quality without coordinate superposition, (3) consistency multiple alignments, (4) classification classification. Comparisons other state-of-the-art (TM-align, FATCAT, DaliLite) using our benchmark dataset showed performed robustly obtaining high-quality alignments generating consistent high coverage accuracy rates, it extremely well when discriminating nonhomologous both single multi-domain comparisons. software freely available academic users stand-alone http://www.pharm.kitasato-u.ac.jp/bmd/bmd/Publications.html.