AlphaFold-SFA: accelerated sampling ofcryptic pocketopening,protein-ligandbinding andallosteryby AlphaFold, slow feature analysis and metadynamics

Abstract

Samplingrare eventsin proteins is crucial for comprehending complex phenomena like cryptic pocket opening, where transient structural changes expose new binding sites. Understanding these rare events also sheds light on protein-ligand binding and allosteric communications, where distant site interactions influence protein function. Traditional unbiased molecular dynamics simulations often fail to sample such rare events, as the free energy barrier between metastable states is large relative to the thermal energy. This renders these events inaccessible on the timescales typically simulated by standard molecular dynamics, limiting our understanding of these critical processes. In this working paper, we proposed a novel unsupervised learning approach termed asslow feature analysis(SFA) which aims to extract slowly varying features from high-dimensional temporal data. SFA trained on small unbiased molecular dynamics simulations launched from AlphaFold generated conformational ensembles manages to capture rare events governing cryptic pocket opening in plasmepsin II. Metadynamics simulations using SFA as collective variables manage to sample ‘deep’ cryptic pocket opening within a few hundreds of nanoseconds which was beyond the reach of microsecond long unbiased molecular dynamics simulations. Taken together, our results show how SFA acts as a dimensionality reduction tool which bridges the gap between AlphaFold, molecular dynamics simulation and metadynamics in context of capturing rare events in biomolecules, extending the scope of structure-based drug discovery in the era of AlphaFold.

WITHDRAWN: Revealing Mechanism of Allostery in RIPK2 kinase

“…Comparing molecular simulations seeded by AlphaFold with unbiased molecular dynamics simulations of the RIPK2-XIAP complex shows that apo RIPK2 fluctuates between active and inactive states, while XIAP binding stabilizes RIPK2's active state. This finding corroborates the hypothesis by Vats et al 10 , underscoring the importance of the DFG loop, activation loop, and αC helix in RIPK2's conformational transitions.…”

“…Demarcated lines separate active from the inactive ensemble of RIPK2. Unbiased molecular dynamics simulation of RIPK2+XIAP complex was reported previously by Vats and coworkers 10 .…”

“…This finding corroborates the hypothesis by Vats et al . 10 , underscoring the importance of the DFG loop, activation loop, and αC helix in RIPK2’s conformational transitions.…”

“…Slow feature analysis (SFA) 10,11 is a dimensionality reduction method, aiming to identify slowly varying features from high-dimensional timeseries data produced by molecular dynamics simulations. SFA seeks to find the most slowly varying features, under the assumption that these slow-varying features represent rare events in biology which plays critical role in conformational dynamics in biomolecules and protein-ligand binding.…”

“…This gap in structural data hinders the understanding of the activation loop's function in allosteric signaling between the kinase orthosteric site and the C-lobe domain 9 . Our recent work utilized AlphaFold, molecular dynamics simulations, and metadynamics to explore the role of the activation loop in the conformational dynamics of RIPK2 10 . We demonstrated that significant conformational changes at the orthosteric site, the activation loop, and the C-lobe domain are crucial for the transition between RIPK2's active and inactive states (Figure 2).…”