SH3-SPOT: an algorithm to predict preferred ligands to different members of the SH3 gene family.

摘要: We have developed a procedure to predict the peptide binding specificity of an SH3 domain from its sequence. The utilizes information extracted position-specific contacts derived six SH3/peptide or SH3/protein complexes known structure. framework defined on structure is used build residue-residue interaction database ligands obtained by panning libraries displayed filamentous phage. SH3-specific multidimensional array containing frequencies contacts. As input, SH3-SPOT requires sequence and query decapeptide ligand. array, that we call matrix, then evaluate probability would bind given domain. This fast enough be applied entire protein database. Panning experiments were performed search putative specific different domains in decapeptides, sequences. ranked some natural partners number among best approximately 5. 6x10(9) decapeptides SWISSPROT expect predictive power method increase with enrichment matrix data new complex structures characterization ligands. was using family as test case but application can easily extended other families (such as, SH2, MHC, EH, PDZ, etc.).