Estimating local backbone structural deviation in homology models

After the atomic coordinates themselves, the most important data in a homol ogy model are the spatial reliability estimates associated with each of the atoms (atom annotation). Recent blind homology modeling predictions have d emonstrated that principally correct sequence-structure alignments are achi evable to sequence identities as low as 25% [Martin, A.C., MacArthur, M.W., Thornton, J.M., 1997. Assessment of comparative modeling in CASP2. Protein s Suppl(1), 14-28]. The locations and extent of spatial deviations in the b ackbone between correctly aligned homologous protein structures remained ve ry poorly estimated however, and these errors were the cause of errant loop predictions [Abagyan, R., Batalov, S., Cardozo, T., Totrov, M., Webber, J. , Zhou, Y., 1997. Homology modeling with internal coordinate mechanics defo rmation zone mapping and improvements of models via conformational search. Proteins Suppl(1), 29-37]. In order to derive accurate measures for local b ackbone deviations, we made a systematic study of static local backbone dev iations between homologous pairs of protein structures. We found that 'thro ugh space' proximity to gaps and chain termini, local three-dimensional 'de nsity', three-dimensional environment conservation, and B-factor of the tem plate contribute to local deviations in the backbone in addition to local s equence identity. Based on these finding, we have identified the meaningful ranges of values within which each of these parameters correlates with sta tic local backbone deviation and produced a combined scoring function to gr eatly improve the estimation of local backbone deviations. The optimized fu nction has more than twice the accuracy of local sequence identity or B-fac tor alone and was validated in a recent blind structure prediction experime nt. This method may be used to evaluate the utility of a preliminary homolo gy model for a particular biological investigation (e.g. drug design) or to provide an improved starting point for molecular mechanics loop prediction methods. (C) 2000 Elsevier Science Ltd. All rights reserved.