Coupling Quantum Mechanical Modeling and Molecular Dynamics on Heterogeneous Supercomputers for Studying Distal Mutation Effects on Drug Binding in HIV-1

Abstract

Predicting how protein mutations affect drug binding remains a major challenge, particularly when the mutations are distal from the binding site. In this study, we introduce a coupled simulation workflow that combines long-time-scale molecular dynamics (MD) with high-throughput quantum mechanical (QM) analysis to reveal the electronic structure signatures of mutation induced drug resistance in the HIV-1 protease. Our workflow leverages GPU-accelerated MD to generate conformational ensembles, and performs in-operando linear-scaling density functional theory (DFT) calculations on selected frames parallelized on a coupled partition of CPU nodes. This design enables efficient, massively parallel quantum analysis of protein-ligand complexes at atomic resolution. Using this approach, we investigate resistance to the antiviral Darunavir in a multi-mutant HIV-1 protease variant. By mapping the network of electronic interactions across the binding interface, our results highlight the critical role of conformational sampling and quantum insight in understanding distal mutation effects, and demonstrate a scalable computational strategy for studying complex biophysical mechanisms of drug resistance. We argue that such kind of analysis may pave the way for designing inhibitors that maintain binding stability against systemic, mutation-induced destabilization.

Figures (11)

• Figure 1: Mutation sites of the HIV-1 PR. The protein is represented in the cartoon style and the ligand with licorice. The blue residues are the catalytic triad (D25, T26 and G27), red residues nearby mutations, and yellow residues the distal mutations.
• Figure 2: Overall coupling of the MD + QM + QM-CR workflow through remotemanager (rm). Four different datasets are generated with remotemanager for running GENESIS (MM), extracting snapshots, running BigDFT (QM), and serialization (QM-CR). Each dataset runs its job as soon as data becomes available.
• Figure 3: Timeline of calculations for the first production phase (40 ns) of the wild type variant. As the GENESIS calculation runs, snapshots are produced, which are transformed into BigDFT and serialization jobs.
• Figure 4: Fragmentation of DRV using the QM-CR purity indicator. The top right table assigns an abbreviation to each fragment, describes the fragments in terms of chemical groups, and reports the purity indicator ($|\Pi|$) from a gas phase calculation. The Purity Indicator identifies the independent modular elements through which DRV works. Therefore, it allows us to identify which specific chemical groups may be affect by mutation-induced loss of binding. The orange dashed lines represent the additional fragmentation that would emerge when using a looser cutoff value of $0.1$.
• Figure 5: Distribution of the purity values of different inhibitor fragments (DRV), protein amino acids, and water molecules over the course of the production runs.