作者: Miguel M. F. Bugalho , Arlindo L. Oliveira
DOI: 10.1007/978-3-540-85557-6_3
关键词:
摘要: Discrete models are important to reduce the complexity of protein folding problem. However, a compromise must be made between model and accuracy model. Previous work by Park Levitt has shown that backbone can modeled with good four state discrete models. Nonetheless, for ab-initio folding, side chains determine if structure is physically possible well packed. We extend taking into account positioning chain in evaluation accuracy. We show problem becomes much harder more dependent on type being modeled. In fact, fitting method used their no longer adequate this extended version propose new test accuracy. The presented results that, some proteins, cannot achieve only Nevertheless, majority proteins an RMSD angstrom or less obtained, and, many those, we reach near two limit. These prove obtain low resolution Since already present models, refinement these solutions simpler effective.