Fundamental Challenges and Outlook in Simulating Liquid–Liquid Phase Separation of Intrinsically Disordered Proteins

A Time-Resolved Diffusion Technique for Detection of the Conformational Changes and Molecular Assembly/Disassembly Processes of Biomolecules

“…Although these methods are useful to study LLPS on the molecular basis, the structural analyses are restricted to the equilibrium states due to their low time resolutions. Other methods such as small angle X-ray scattering (SAXS), atomic force microscopy (AFM), and fluorescence resonance energy transfer (FRET) have been used to investigate the conformational heterogeneities of IDRs ( Mukhopadhyay, 2020 ; Surewicz and Babinchak, 2020 ; Bari and Prakashchand, 2021 ; Kodera et al, 2021 ). However, there are several difficulties to study the LLPS processes by these methods, although the time resolutions of them have been improved.…”

Modelling the structure and interactions of intrinsically disordered peptides with multiple-replica, metadynamics-based sampling methods and force-field combinations

“…Molecular dynamics (MD) simulations can in principle offer the opportunity of resolving structural, kinetic and thermodynamic characteristics of biological systems such as IDPs and IDP-rich bodies at the length and time scales that are not easily accessible experimentally 22,[37][38][39][40][41][42][43][44][45][46][47][48][49][50][51] . These simulations complement state-of-the-art experimental methods such as NMR spectroscopy [52][53] , single-molecule Förster resonance energy transfer (sm-FRET) [54][55] , and small-angle X-ray and neutron scattering (SAXS and SANS) [56][57] , which may exhibit challenges in measuring molecular motions at atomic resolution and conformational heterogeneity associated with structural disorder 37,[41][42]49 . Nevertheless, conformational sampling in classical molecular dynamics (MD) simulations can be restrained to local minima of free energy surfaces (FES), and accessing the full and complex landscape of disordered proteins can be nontrivial with standard unbiased simulations.…”

“…It is well-known that the results of MD simulations can strongly depend strongly on the accuracy of the applied protein-water force-fields, which may lead to large discrepancies with various experimental measurements 42,45,49,[74][75][76][77][78][79][80][81][82] . Examples have been observed with state-of-the-art Amber force-fields a99SB*-ILDN [83][84] To study more accurately the structural motifs emerging from the extensive sampling of DHH1N conformational ensemble, we also evaluate key secondarystructure content for CHARMM36m and CHARMM22* PTMetaD-WTE by means of the DSSP algorithm 98 .…”

Modelling the structure and interactions of intrinsically disordered proteins with multiple-replica, metadynamics-based sampling methods and force-field combinations

“…Molecular dynamics (MD) simulations can in principle offer the opportunity of resolving structural, kinetic and thermodynamic characteristics of biological systems such as IDPs and IDP-rich bodies at the length and time scales that are not easily accessible experimentally 23,[38][39][40][41][42][43][44][45][46][47][48][49][50][51][52] . Recently, many protein folding problems were successfully solved by the advance of the deep learning method AlphaFold 53 .…”

“…Thus, exploring the conformations of IDPs still largely rely on molecular dynamic simulations 16 . MD simulations complement state-of-the-art experimental methods such as NMR spectroscopy [54][55] , single-molecule Förster resonance energy transfer (sm-FRET) [56][57] , and small-angle X-ray and neutron scattering (SAXS and SANS) [58][59] , which may exhibit challenges in measuring molecular motions at atomic resolution and conformational heterogeneity associated with structural disorder 38,[42][43]50 . Nevertheless, conformational sampling in classical molecular dynamics (MD) simulations can be restrained to local minima of free energy surfaces (FES), and accessing the full and complex landscape of disordered proteins can be nontrivial with standard unbiased simulations.…”