Predicting peptide interactions with model class II MHC structures

Artikel i vetenskaplig tidskrift, 2005

An automated method for constructing 3D models of class II MHC structures that uses constraint logic programming to select side-chain conformations is described. This method follows a comparative modeling approach in basing the model structures on experimentally determined MHC-peptide structures, but it uses constraints to ease open the peptide binding groove so that the modeled MHC structure is a less specific fit for the co-crystallized peptide in the starting structure. The resulting models are used by a "peptide threading" program that attempts to predict peptides from a protein sequence that will bind strongly to particular MHC alleles. Our results indicate that MHC models that have been constructed in this way enable the peptide threading program to make binding predictions that are comparable with those obtained when using experimentally determined MHC structures, suggesting that a combined modeling and peptide threading approach is worth pursuing for MHC molecules for which experimentally determined structures are not available.