A molecular dynamics and knowledge-based computational strategy to predict native-like structures of polypeptides

摘要: One of the main research problems in structural bioinformatics is prediction three-dimensional structures (3-D) polypeptides or proteins. The current rate at which amino acid sequences are identified increases much faster than 3-D protein structure determination by experimental methods, such as X-ray diffraction and NMR techniques. both experimentally expensive time consuming. Predicting correct a molecule an intricate arduous task. (PSP) problem is, computational complexity theory, NP-complete problem. In order to reduce computing time, efforts have targeted hybridizations between ab initio knowledge-based methods aiming efficient polypeptides. this article we present hybrid method for An artificial neural network that predicts approximated combined with strategy. Molecular dynamics (MD) simulation used refinement structures. step, global interactions each pair atoms (including non-bond interactions) evaluated. developed MD protocol enables us polypeptide torsion angles deviation from predicted improve their stereo-chemical quality. obtained results shows predict native-like considerably reduced. We test our strategy four mini proteins whose sizes vary 19 34 residues. end 32.0nanoseconds (ns) were comparable topologically correspondent