Database algorithm for generating protein backbone and side-chain co-ordinates from a C alpha trace application to model building and detection of co-ordinate errors.

摘要: Abstract The problem of constructing all-atom model co-ordinates a protein from an outline the polypeptide chain is encountered in structure determination by crystallography or nuclear magnetic resonance spectroscopy, building homology and design. Here, we present automatic procedure for generating full (backbone and, optionally, side-chains) given Cα trace amino acid sequence. To construct backbones, database first scanned fragments that locally fit according to distance criteria. A best path algorithm then sifts through these segments selects optimal with minimal mismatch at fragment joints. In blind tests, using fully known structures, backbones (Cα, C, N, O) can be reconstructed reliability 0.4 0.6 root-mean-square position deviation not more than 0 5% peptide flips. This accuracy sufficient identify possible errors co-ordinate sets. co-ordinates, side-chains are added library frequently occurring rotamers simple fast Monte Carlo simulated annealing. tests on X-ray structures determined better 2.5 resolution, positions side-chain atoms core (less 20% relative accessibility) have 1.6 (r.m.s. deviation) 70% χ1 angles within 30 ° structure. computer program MaxSprout available request.