#compchem #quantumcomputing New paper published in Physical Review research: "Practical protein-pocket hydration-site prediction for drug discovery on a quantum computer". Great collaboration with @qubit_pharma and @qctrlHQ ! Here are the key takeaways from this work: 🔬Formulating the Problem: We mapped the 3D hydration-site prediction task onto a Quadratic Unconstrained Binary Optimization (QUBO) problem by fitting 3D Reference Interaction Site Model (3D-RISM) continuous water densities into a Gaussian Mixture Model (GMM). ⚡ Demonstrating Scale on Real Quantum Hardware:Ran hardware experiments on 156-qubit IBM Heron processors using Q-CTRL’s Fire Opal optimization solver. Successfully scaled up to 123 qubits to solve 3D water placement in real-life protein-ligand complexes (FDA-approved drug target sites). Matched the accuracy of classical approaches while achieving significantly higher success probabilities than traditional simulated annealing (SA) heuristics on hardware. 📈 Path to Quantum Utility: Detailed resource estimations reveal that accuracy systematically improves as the problem scale expands toward ~900+ qubits. With rapid developments in hardware and early error correction, full quantum advantage for CADD setup tasks, such as molecular docking and lead optimization, is approaching. A massive thank you to co-authors: @loco_daniele, Kisa Barkemeyer, Andre R. R. Carvalho for driving this forward! 📄 Read the full paper here: doi.org/10.1103/gyqr-mvlh

New paper, published #openaccess in Digital Discovery: "Real‑Space Chemistry on Quantum Computers: A Fault‑Tolerant Algorithm with Adaptive Grids and Transcorrelated Extension". First-quantized, real-space formulations of quantum chemistry on quantum computers are appealing: qubit count scales logarithmically with spatial resolution, and Coulomb operators achieve quadratic instead of quartic computational scaling of two-electron interactions. However, existing schemes employ uniform discretizations, so the resolution required to capture electron–nuclear cusps in high-density regions oversamples low-density regions, wasting computational resources. We address this by deploying non-uniform, molecule-adaptive grids that concentrate points where electronic density is high. Using Voronoi partitions of these grids, the molecular Hamiltonian is expressed in a Hermitian form and in a transcorrelated, isospectral form that eliminates Coulomb singularities and yields cusp-free eigenfunctions. Both formulations slot naturally into quantum eigenvalue solvers: Hermitian Quantum Phase Estimation (#QPE) and the recent generalised Quantum Eigenvalue Estimation (#QEVE) protocol for its non-Hermitian, transcorrelated counterpart. Numerical validation on benchmark systems confirms this ab–initio framework paves a promising route to ground-state chemistry on quantum hardware. Great work from @ChemCesar and nice collaboration with E. Giner. #compchem #quantumcomputing @qubit_pharma doi.org/10.1039/D6DD00002A

Now in final form and #openaccess. "Quantum circuits for the Metropolis–Hastings algorithm" #quantumcomputing iopscience.iop.org/article/1…

𝐁𝐫𝐞𝐚𝐤𝐢𝐧𝐠 𝐭𝐡𝐞 𝐒𝐩𝐞𝐞𝐝 𝐋𝐢𝐦𝐢𝐭 𝐢𝐧 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐌𝐨𝐥𝐞𝐜𝐮𝐥𝐚𝐫 𝐃𝐲𝐧𝐚𝐦𝐢𝐜𝐬! I am thrilled to share our latest paper published in the Journal of Chemical Theory and Computation (JCTC) @JCIM_JCTC : "𝐅𝐚𝐬𝐭𝐞𝐫 𝐌𝐨𝐥𝐞𝐜𝐮𝐥𝐚𝐫-𝐃𝐲𝐧𝐚𝐦𝐢𝐜𝐬 𝐰𝐢𝐭𝐡 𝐍𝐞𝐮𝐫𝐚𝐥 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐏𝐨𝐭𝐞𝐧𝐭𝐢𝐚𝐥𝐬 𝐯𝐢𝐚 𝐃𝐢𝐬𝐭𝐢𝐥𝐥𝐞𝐝 𝐌𝐮𝐥𝐭𝐢𝐩𝐥𝐞 𝐓𝐢𝐦𝐞-𝐬𝐭𝐞𝐩𝐩𝐢𝐧𝐠 𝐚𝐧𝐝 𝐍𝐨𝐧𝐜𝐨𝐧𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐯𝐞 𝐅𝐨𝐫𝐜𝐞𝐬". Link to the paper: pubs.acs.org/doi/10.1021/acs… Link to the updated preprint (#openaccess): arxiv.org/abs/2602.14975 In this work, we introduce DMTS-NC, a model-agnostic acceleration strategy that bridges the performance gap between high-accuracy Neural Network Potentials (NNPs) and traditional classical force fields. By distilling large foundation models (like FeNNix-Bio1 and MACE-OFF23) into a lightweight architecture that directly predicts nonconservative forces, we bypass the computationally expensive backpropagation step entirely. To ensure exceptional simulation stability without requiring tedious, system-specific fine-tuning, the architecture strictly enforces key physical priors like rotational equivariance and net-zero atomic force cancellation. The results are a game-changer for high-throughput simulations: DMTS-NC delivers a 15% 𝐭𝐨 30% 𝐬𝐩𝐞𝐞𝐝𝐮𝐩 𝐨𝐯𝐞𝐫 𝐩𝐫𝐞𝐯𝐢𝐨𝐮𝐬 𝐜𝐨𝐧𝐬𝐞𝐫𝐯𝐚𝐭𝐢𝐯𝐞 𝐌𝐓𝐒 𝐟𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤𝐬 𝐚𝐧𝐝 𝐮𝐩 𝐭𝐨 𝐚 5.6𝐱 𝐚𝐜𝐜𝐞𝐥𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐯𝐞𝐫 𝐬𝐢𝐧𝐠𝐥𝐞-𝐭𝐢𝐦𝐞-𝐬𝐭𝐞𝐩 𝐦𝐞𝐭𝐡𝐨𝐝𝐬. Furthermore, by pairing this approach with Hydrogen Mass Repartitioning (HMR) and High Hydrogen Friction (HHF), we successfully extended outer time steps up to 10 fs while maintaining flawless thermodynamic and free-energy accuracy. Implemented in Tinker-HP @TINKERtoolsMD ! Huge congratulations to the team at @qubit_pharma and Sorbonne Université/CNRS: @NicolaiGouraud, @comecattin, Olivier ADJOUA, Louis Lagardère, and @Thomas__Ple ! Funding from @ERC_Research(project EMC2) and supercomputing time @Genci_fr . #MolecularDynamics #MachineLearning #compchem #compbio #AI4Science #foundationmodels #FeNNixBio1

New group preprint: "Faster Molecular Dynamics with Neural Network Potentials via Distilled Multiple Time-Stepping and Non-Conservative Forces”, which introduces the DMTS-NC approach - a strategy designed to further accelerate atomistic molecular dynamics simulations using foundation neural network models. arxiv.org/abs/2602.14975 Significant Speedups: using the FeNNix-Bio1 foundation model, DMTS-NC is 15–30% faster than its conservative DMTS counterpart making it 3 to 4 time faster overall than single-stime-step (STS) .When applied to the somewhat more computationnaly MACE-OFF23, we observed speedups ranging from 3.66 to 5.64 compared to STS. No Fine-Tuning and Enhanced Stability: unlike many acceleration methods, DMTS-NC requires no system-specific fine-tuning, making it much easier to implement across different chemical spaces. Pushing the Limits: by combining Hydrogen Mass Repartitioning (HMR) and High Hydrogen Friction (HHF), we successfully extended external timesteps up to 10fs while maintaining accuracy. This framework is model-agnostic and can be applied to any neural network potential. We’ve validated it on systems ranging from bulk water to solvated proteins like DHFR. For those interested in the technical details or implementation, the code is available via the FeNNol library and Tinker-HP. Great work by @NicolaiGouraud @qubit_pharma #MachineLearning #compchem #Biophysics #DrugDesign #AI

#compchem Despite #AI, characterizing dynamic interactions in realistic biological environments remains a massive computational bottleneck in modern #drugdiscovery. This perspective identifies the integration of #HPC, #machinelearning & #QuantumComputing as a definitive strategy to adress this issue. 🔬 𝐂𝐨𝐫𝐞 𝐒𝐜𝐢𝐞𝐧𝐭𝐢𝐟𝐢𝐜 𝐂𝐨𝐧𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧𝐬 ▸𝐅𝐞𝐍𝐍𝐢𝐱-𝐁𝐢𝐨𝐥 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥: A family of Machine Learning models engineered for high-speed, reactive atomistic #moleculardynamics simulations, achieving quantum-level precision with #GPU-accelerated scalability. ▸𝐇𝐢𝐠𝐡-𝐀𝐜𝐜𝐮𝐫𝐚𝐜𝐲 𝐐𝐮𝐚𝐧𝐭𝐮𝐦 𝐀𝐥𝐠𝐨𝐫𝐢𝐭𝐡𝐦𝐬: we introduce new approaches to extent the 𝐈𝐠𝐧𝐢𝐬 quantum chemistry database used for FeNNix-Bio: from improved VQE for #NISQ to innovative #FTQC building blocks. Our hybrid methodology utilizes a classical embedding "shortcut" to reach the Complete-Basis-Set limit (CBS). Using less qubits, we achieve chemical accuracy that would typically require 300+ qubits, up to a 10-fold reduction in resource overhead. ▸𝐁𝐫𝐢𝐝𝐠𝐢𝐧𝐠 𝐭𝐡𝐞 𝐓𝐢𝐦𝐞𝐬𝐜𝐚𝐥𝐞 𝐆𝐚𝐩: 𝐂𝐥𝐚𝐬𝐬𝐢𝐜𝐚𝐥 𝐄𝐧𝐡𝐚𝐧𝐜𝐞𝐝 𝐒𝐚𝐦𝐩𝐥𝐢𝐧𝐠. We introduce accelerators for drug binding in proteins: Lambda-ABF-OPES & Dual-LAO for absolute/relative free energies of binding, provide up to a 30-fold speedup over standard protocols. 𝐐𝐮𝐚𝐧𝐭𝐮𝐦-𝐄𝐧𝐡𝐚𝐧𝐜𝐞𝐝 𝐒𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐚𝐥 𝐌𝐞𝐜𝐡𝐚𝐧𝐢𝐜𝐬. We leverage quadratic speedups for sampling & highlight an experimental milestone: the realization of quantum Markov Chain Monte Carlo on Quantinuum’s H2/Helios #QPU. ▸📷 𝐓𝐡𝐞 𝐇𝐲𝐩𝐞𝐫𝐢𝐨𝐧 𝐐𝐮𝐚𝐧𝐭𝐮𝐦 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦 Our GPU-accelerated emulator/orchestrator enables custom kernels, specifically engineered for efficient sparse matrix–sparse vector multiplication & use of NVIDIA cuTensor for: - 𝐒𝐮𝐩𝐞𝐫 𝐅𝐚𝐬𝐭 𝐄𝐱𝐚𝐜𝐭 𝐒𝐭𝐚𝐭𝐞-𝐕𝐞𝐜𝐭𝐨𝐫 & 𝐌𝐚𝐭𝐫𝐢𝐱 𝐏𝐫𝐨𝐝𝐮𝐜𝐭 𝐒𝐭𝐚𝐭𝐞𝐬 (𝐌𝐏𝐒) 𝐄𝐦𝐮𝐥𝐚𝐭𝐢𝐨𝐧: 30++ logical qubits & 10,000++ CNOT gates on GPUs enabling to efficiently trigger VQE/QPE at full accuracy. - 𝐇𝐚𝐫𝐝𝐰𝐚𝐫𝐞-𝐀𝐠𝐧𝐨𝐬𝐭𝐢𝐜 𝐐𝐏𝐔 𝐎𝐫𝐜𝐡𝐞𝐬𝐭𝐫𝐚𝐭𝐢𝐨𝐧: real-life benchmarks across IBM Heron, Quantinuum H2/Helios, IonQ Aria/Forte & Pasqal technologies. Thanks to all co-authors including those on X: @Narjes_Ansari @ChemCesar @loco_daniele @FefeAviat @blazhynska66497 @Dr_DiataTraore @NicolaiGouraud @Thomas__Ple

#compchem Our paper in J. Phys. Chem. Lett. (@JPhysChem): "Accelerating Molecular Dynamics Simulations with Foundation Neural Network Models using Multiple Time-Step and Distillation" made it to one of the covers! pubs.acs.org/doi/full/10.102…