Neural Upscaling from Residue-Level Protein Structure Networks to Atomistic Structures

• Published in: Biomolecules (Basel, Switzerland) Switzerland), 2021-11, Vol.11 (12), p.1788
• Main Authors: Duong, Vy T, Diessner, Elizabeth M, Grazioli, Gianmarc, Martin, Rachel W, Butts, Carter T
• Format: Article
• Language: English
• Subjects: Alzheimer's disease; coarse-grained models; Computer applications; Cost control; intrinsically disordered proteins; Intrinsically Disordered Proteins - chemistry; Learning algorithms; Machine Learning; Macromolecules; Models, Molecular; molecular dynamics; Molecular Dynamics Simulation; Neural networks; Neural Networks, Computer; Parameter estimation; Protein structure; protein structure networks; Proteins; Secondary structure; Simulation; Thermodynamics
• ISSN: 2218-273X; 2218-273X
• DOI: 10.3390/biom11121788

Coarse-graining is a powerful tool for extending the reach of dynamic models of proteins and other biological macromolecules. Topological coarse-graining, in which biomolecules or sets thereof are represented via graph structures, is a particularly useful way of obtaining highly compressed representations of molecular structures, and simulations operating via such representations can achieve substantial computational savings. A drawback of coarse-graining, however, is the loss of atomistic detail-an effect that is especially acute for topological representations such as protein structure networks (PSNs). Here, we introduce an approach based on a combination of machine learning and physically-guided refinement for inferring atomic coordinates from PSNs. This "neural upscaling" procedure exploits the constraints implied by PSNs on possible configurations, as well as differences in the likelihood of observing different configurations with the same PSN. Using a 1 μs atomistic molecular dynamics trajectory of Aβ1-40, we show that neural upscaling is able to effectively recapitulate detailed structural information for intrinsically disordered proteins, being particularly successful in recovering features such as transient secondary structure. These results suggest that scalable network-based models for protein structure and dynamics may be used in settings where atomistic detail is desired, with upscaling employed to impute atomic coordinates from PSNs.