GNMCADS: Sampling For Protein Conformation Diversity With Gaussian Network Model Guided Condition Annealed Diffusion Sampler
Proteins are dynamic molecules existing in diverse conformational states underlying their biological functions. Although recent approaches have enabled diverse conformational sampling by emulating molecular dynamics simulations, perturbing evolutionary information, or steering internal mechanisms of structure prediction models, predicting conformations resulting from major domain motions or…
Proteins are intricate molecules with various conformational states that determine their biological functions. Current methods have facilitated diverse conformational sampling using molecular dynamics simulations, evolutionary perturbations, and internal structure prediction mechanisms. However, predicting conformations arising from significant domain motions or long timescale movements remains a challenge.
In response to this issue, we present GNMCADS, a novel conformational sampling strategy that augments the diversity of protein diffusion models by strategically annealing the conditioning signal based on the inherent dynamical organization of the protein being sampled. Furthermore, we integrate GNMCADS into the diffusion module of AlphaFold3, allowing for the generation of varied protein conformations.
In comparative tests involving 92 proteins, encompassing 54 class A GPCRs, 15 transporters, and 23 proteins characterized by major domain movements, GNMCADS demonstrated superior diversity in sampling when compared to alternative state-of-the-art conformational sampling techniques.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
