Author: Lim Heo; Michael Feig
Title: Modeling of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Proteins by Machine Learning and Physics-Based Refinement Document date: 2020_3_28
ID: 9qv11m4f_8
Snippet: When we applied our latest refinement protocol to AlphaFold protein models, the structure changes upon refinement were moderate, usually less than 2 Å in Cɑ-RMSD. ( Table 2 ) The most significant changes occurred mostly in loops and the relative orientation between secondary structure elements. For example, in Figure 1D , a loop structure was changed, and the relative orientations of an α-helix and two β-strands were adjusted with respect to.....
Document: When we applied our latest refinement protocol to AlphaFold protein models, the structure changes upon refinement were moderate, usually less than 2 Å in Cɑ-RMSD. ( Table 2 ) The most significant changes occurred mostly in loops and the relative orientation between secondary structure elements. For example, in Figure 1D , a loop structure was changed, and the relative orientations of an α-helix and two β-strands were adjusted with respect to the other secondary structure elements. As another example, in Figure 2D , an α-helix was moved to improve hydrophobic packing and salt bridges between charged residues. We found earlier that machine learning-based models can be improved at the atomistic level by physics-based refinement and we expect that to the extent that the general features of the initial models are correct, refinement resulted in improved models.
Search related documents:
Co phrase search for related documents- improved model and initial model: 1, 2, 3, 4, 5
- improved model and loop structure: 1
- learning base and machine learning base: 1, 2, 3, 4, 5, 6
Co phrase search for related documents, hyperlinks ordered by date