GLOW: A Workflow Integrating Gaussian-Accelerated Molecular Dynamics and Deep Learning for Free Energy Profiling

Abstract

We introduce a Gaussian-accelerated molecular dynamics (GaMD), deep learning (DL), and free energy profiling workflow (GLOW) to predict molecular determinants and map free energy landscapes of biomolecules. All-atom GaMD-enhanced sampling simulations are first performed on biomolecules of interest. Structural contact maps are then calculated from GaMD simulation frames and transformed into images for building DL models using a convolutional neural network. Important structural contacts are further determined from DL models of attention maps of the structural contact gradients, which allow us to identify the system reaction coordinates. Finally, free energy profiles are calculated for the selected reaction coordinates through energetic reweighting of the GaMD simulations. We have also successfully demonstrated GLOW for the characterization of activation and allosteric modulation of a G protein-coupled receptor, using the adenosine A 1 receptor (A 1 AR) as a model system. GLOW findings are highly consistent with previous experimental and computational studies of the A 1 AR, while also providing further mechanistic insights into the receptor function. In summary, GLOW provides a systematic approach to mapping free energy landscapes of biomolecules. The GLOW workflow and its user manual can be downloaded at http://miaolab.org/GLOW.

Challenges and frontiers of computational modelling of biomolecular recognition

“…Recently, we have integrated the GaMD, Deep Learning and free energy prOfiling Workflow (GLOW) to predict important reaction coordinates and map free energy profiles of biomolecules (Do et al, 2022). First, GaMD simulations are performed on the target biomolecules (Fig.…”

“…Finally, the free energy profiles of these reaction coordinates are calculated through reweighting of GaMD simulations to characterize the biomolecular systems of interest (Fig. 4D) (Do et al, 2022). GLOW was successfully demonstrated on characterization of activation and allosteric modulation of a GPCR, using the adenosine A1 receptor (A1AR) as a model system.…”

“…GLOW revealed that binding of a PAM (MIPS521) to the agonist-Gi-A1AR complex biased the receptor conformational ensemble, especially in the ECL1 and ECL2 regions. PAM binding stabilized agonist binding within the orthosteric pocket of A1AR, which confined the extracellular mouth of the receptor Furthermore, PAM binding disrupted the N148 ECL2 -V152 ECL2 -helical hydrogen bond and distorted this portion of the ECL2 helix (Do et al, 2022).…”

Dynamic allosteric networks drive adenosine A1receptor activation and G-protein coupling

“…[18] Computational and mutagenesis studies have reported interesting insights about the A 1 R activation and allosteric modulation identifying important residues for the signaling efficacy of agonists and cooperativity of PAMs. [18, 22, 23] Despite all these achievements, a detailed characterization of the allosteric networks that drive receptor activation and G-proteins binding is still missing.…”

Effects of Presenilin-1 Familial Alzheimer’s Disease Mutations on γ-Secretase Activation for Cleavage of Amyloid Precursor Protein

“…In addition, "selective GaMD" algorithms, including Ligand GaMD (LiGaMD) 22 , Peptide GaMD (Pep-GaMD) 23 , and Protein-Protein Interaction-GaMD (PPI-GaMD) 24 have been developed to enable repetitive binding and dissociation of small-molecule ligands, highly exible peptides, and proteins within microsecond simulations, which allow for highly e cient and accurate calculations of ligand/peptide/protein binding free energy and kinetic rate constants 19 . Recently, GaMD has been combined with deep learning and free energy pro ling into GLOW to predict molecular determinants and map free energy landscapes of biomolecules 25 .…”