A Side-Chain Positioning program with electron density map constraints based on an exact optimization framework
Résumé
Darwin is a Side-chain positioning program based on MSL MSLLIB and cost function Networks for modeling the corresponding optimization problem . Side-chain positioning is a central component of the protein structure prediction problem and has been the focus of a large body of research. SCP is NP-hard; in fact, they are even inapproximable. In practice, it is tackled by a variety of general search techniques and specialized heuristics. In this work, We investigate a new formulation of the problem as a bi-objectives function based on cost function networks mixing physical energy and electron density. We introduce two novel rounding schemes which allow founding optimal solutions with a theoretical guarantee and the first results on simulated electron density maps.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |