Predicting the &bgr;-helix fold from protein sequence data
RECOMB '01 Proceedings of the fifth annual international conference on Computational biology
Trilogy: discovery of sequence-structure patterns across diverse proteins
Proceedings of the sixth annual international conference on Computational biology
Segmentation conditional random fields (SCRFs): a new approach for protein fold recognition
RECOMB'05 Proceedings of the 9th Annual international conference on Research in Computational Molecular Biology
| Hi-index | 0.00 |
A method is presented that uses β-strand interactions at both the sequence and the atomic level, to predict the beta-structural motifs in protein sequences. A program called Wrap-and-Pack implements this method, and is shown to recognize β-trefoils, an important class of globular β-structures, in the Protein Data Bank with 92% specificity and 92.3% sensitivity in cross-validation. It is demonstrated that Wrap-and-Pack learns each of the ten known SCOP β-trefoil families, when trained primarily on β-structures that are not β-trefoils, together with 3D structures of known β-trefoils from outside the family. Wrap-and-Pack also predicts many proteins of unknown structure to be β-trefoils. The computational method used here may generalize to other β-structures for which strand topology and profiles of residue accessibility are well conserved.