arXiv subjects

Stephan Thaler

Publications and source records attributed to Stephan Thaler.

12 recordsLinked to original sources

arXiv2026

TerraBind: Fast and Accurate Binding Affinity Prediction through Coarse Structural Representations

We present TerraBind, a foundation model for protein-ligand structure and binding affinity prediction that achieves 26-fold faster inference than state-of-the-art methods while improving affinity prediction accuracy by $\sim$20\%. Current deep learning approaches to structure-based drug design rely on expensive all-atom diffusion to generate 3D coordinates, creating inference bottlenecks that render large-scale compound screening computationally intractable. We challenge this paradigm with a critical hypothesis: full all-atom resolution is unnecessary for accurate small molecule pose and binding affinity prediction. TerraBind tests this hypothesis through a coarse pocket-level representation (protein C$_\beta$ atoms and ligand heavy atoms only) within a multimodal architecture combining COATI-3 molecular encodings and ESM-2 protein embeddings that learns rich structural representations, which are used in a diffusion-free optimization module for pose generation and a binding affinity likelihood prediction module. On structure prediction benchmarks (FoldBench, PoseBusters, Runs N' Poses), TerraBind matches diffusion-based baselines in ligand pose accuracy. Crucially, TerraBind outperforms Boltz-2 by $\sim$20\% in Pearson correlation for binding affinity prediction on both a public benchmark (CASP16) and a diverse proprietary dataset (18 biochemical/cell assays). We show that the affinity prediction module also provides well-calibrated affinity uncertainty estimates, addressing a critical gap in reliable compound prioritization for drug discovery. Furthermore, this module enables a continual learning framework and a hedged batch selection strategy that, in simulated drug discovery cycles, achieves 6$\times$ greater affinity improvement of selected molecules over greedy-based approaches.

cs.LG↗

arXiv2025

Boltz-ABFE: Free Energy Perturbation without Crystal Structures

Free energy perturbation (FEP) is considered the gold-standard simulation method for estimating small molecule binding affinity, a quantity of vital importance to drug discovery. The accuracy of FEP critically depends on an accurate model of the protein-ligand complex as an initial condition for the underlying molecular dynamics simulation. This requirement has limited the impact of FEP in earlier stages of the discovery process, where appropriate experimental crystal structures are rarely available. The latest generation of structure prediction models, such as Boltz-2, promise to overcome this limitation by predicting protein-ligand complex structures. In this work, we combine Boltz-2 with our own absolute FEP protocol to build Boltz-ABFE, a robust pipeline for estimating the absolute binding free energies (ABFE) in the absence of experimental crystal structures. We investigate the quality of the structures predicted by Boltz-2, propose automated approaches to improve structures for use in molecular dynamics simulations, and demonstrate the effectiveness of the Boltz-ABFE pipeline for four protein targets from the FEP+ benchmark set. Demonstrating the feasibility of absolute FEP simulations without experimental crystal structures, Boltz-ABFE significantly expands the domain of applicability of FEP, paving the way towards accelerated early-stage drug discovery via accurate, structure-based affinity estimation.

physics.comp-ph↗

arXiv2025

chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulations

Machine learning potentials (MLPs) have advanced rapidly and show great promise to transform molecular dynamics (MD) simulations. However, most existing software tools are tied to specific MLP architectures, lack integration with standard MD packages, or are not parallelizable across GPUs. To address these challenges, we present chemtrain-deploy, a framework that enables model-agnostic deployment of MLPs in LAMMPS. chemtrain-deploy supports any JAX-defined semi-local potential, allowing users to exploit the functionality of LAMMPS and perform large-scale MLP-based MD simulations on multiple GPUs. It achieves state-of-the-art efficiency and scales to systems containing millions of atoms. We validate its performance and scalability using graph neural network architectures, including MACE, Allegro, and PaiNN, applied to a variety of systems, such as liquid-vapor interfaces, crystalline materials, and solvated peptides. Our results highlight the practical utility of chemtrain-deploy for real-world, high-performance simulations and provide guidance for MLP architecture selection and future design.

physics.comp-ph↗

arXiv2025

JaxSGMC: Modular stochastic gradient MCMC in JAX

We present JaxSGMC, an application-agnostic library for stochastic gradient Markov chain Monte Carlo (SG-MCMC) in JAX. SG-MCMC schemes are uncertainty quantification (UQ) methods that scale to large datasets and high-dimensional models, enabling trustworthy neural network predictions via Bayesian deep learning. JaxSGMC implements several state-of-the-art SG-MCMC samplers to promote UQ in deep learning by reducing the barriers of entry for switching from stochastic optimization to SG-MCMC sampling. Additionally, JaxSGMC allows users to build custom samplers from standard SG-MCMC building blocks. Due to this modular structure, we anticipate that JaxSGMC will accelerate research into novel SG-MCMC schemes and facilitate their application across a broad range of domains.

stat.CO↗

arXiv2024

Implicit Delta Learning of High Fidelity Neural Network Potentials

Neural network potentials (NNPs) offer a fast and accurate alternative to ab-initio methods for molecular dynamics (MD) simulations but are hindered by the high cost of training data from high-fidelity Quantum Mechanics (QM) methods. Our work introduces the Implicit Delta Learning (IDLe) method, which reduces the need for high-fidelity QM data by leveraging cheaper semi-empirical QM computations without compromising NNP accuracy or inference cost. IDLe employs an end-to-end multi-task architecture with fidelity-specific heads that decode energies based on a shared latent representation of the input atomistic system. In various settings, IDLe achieves the same accuracy as single high-fidelity baselines while using up to 50x less high-fidelity data. This result could significantly reduce data generation cost and consequently enhance accuracy and generalization, and expand chemical coverage for NNPs, advancing MD simulations for material science and drug discovery. Additionally, we provide a novel set of 11 million semi-empirical QM calculations to support future multi-fidelity NNP modeling.

physics.chem-ph↗

arXiv2024

OpenQDC: Open Quantum Data Commons

Machine Learning Interatomic Potentials (MLIPs) are a highly promising alternative to force-fields for molecular dynamics (MD) simulations, offering precise and rapid energy and force calculations. However, Quantum-Mechanical (QM) datasets, crucial for MLIPs, are fragmented across various repositories, hindering accessibility and model development. We introduce the openQDC package, consolidating 37 QM datasets from over 250 quantum methods and 400 million geometries into a single, accessible resource. These datasets are meticulously preprocessed, and standardized for MLIP training, covering a wide range of chemical elements and interactions relevant in organic chemistry. OpenQDC includes tools for normalization and integration, easily accessible via Python. Experiments with well-known architectures like SchNet, TorchMD-Net, and DimeNet reveal challenges for those architectures and constitute a leaderboard to accelerate benchmarking and guide novel algorithms development. Continuously adding datasets to OpenQDC will democratize QM dataset access, foster more collaboration and innovation, enhance MLIP development, and support their adoption in the MD field.

physics.chem-ph↗

arXiv2024

ET-Flow: Equivariant Flow-Matching for Molecular Conformer Generation

Predicting low-energy molecular conformations given a molecular graph is an important but challenging task in computational drug discovery. Existing state-of-the-art approaches either resort to large scale transformer-based models that diffuse over conformer fields, or use computationally expensive methods to generate initial structures and diffuse over torsion angles. In this work, we introduce Equivariant Transformer Flow (ET-Flow). We showcase that a well-designed flow matching approach with equivariance and harmonic prior alleviates the need for complex internal geometry calculations and large architectures, contrary to the prevailing methods in the field. Our approach results in a straightforward and scalable method that directly operates on all-atom coordinates with minimal assumptions. With the advantages of equivariance and flow matching, ET-Flow significantly increases the precision and physical validity of the generated conformers, while being a lighter model and faster at inference. Code is available https://github.com/shenoynikhil/ETFlow.

q-bio.QM↗

arXiv2024

chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physics

Neural Networks (NNs) are effective models for refining the accuracy of molecular dynamics, opening up new fields of application. Typically trained bottom-up, atomistic NN potential models can reach first-principle accuracy, while coarse-grained implicit solvent NN potentials surpass classical continuum solvent models. However, overcoming the limitations of costly generation of accurate reference data and data inefficiency of common bottom-up training demands efficient incorporation of data from many sources. This paper introduces the framework chemtrain to learn sophisticated NN potential models through customizable training routines and advanced training algorithms. These routines can combine multiple top-down and bottom-up algorithms, e.g., to incorporate both experimental and simulation data or pre-train potentials with less costly algorithms. chemtrain provides an object-oriented high-level interface to simplify the creation of custom routines. On the lower level, chemtrain relies on JAX to compute gradients and scale the computations to use available resources. We demonstrate the simplicity and importance of combining multiple algorithms in the examples of parametrizing an all-atomistic model of titanium and a coarse-grained implicit solvent model of alanine dipeptide.

physics.chem-ph↗

arXiv2022

Scalable Bayesian Uncertainty Quantification for Neural Network Potentials: Promise and Pitfalls

Neural network (NN) potentials promise highly accurate molecular dynamics (MD) simulations within the computational complexity of classical MD force fields. However, when applied outside their training domain, NN potential predictions can be inaccurate, increasing the need for Uncertainty Quantification (UQ). Bayesian modeling provides the mathematical framework for UQ, but classical Bayesian methods based on Markov chain Monte Carlo (MCMC) are computationally intractable for NN potentials. By training graph NN potentials for coarse-grained systems of liquid water and alanine dipeptide, we demonstrate here that scalable Bayesian UQ via stochastic gradient MCMC (SG-MCMC) yields reliable uncertainty estimates for MD observables. We show that cold posteriors can reduce the required training data size and that for reliable UQ, multiple Markov chains are needed. Additionally, we find that SG-MCMC and the Deep Ensemble method achieve comparable results, despite shorter training and less hyperparameter tuning of the latter. We show that both methods can capture aleatoric and epistemic uncertainty reliably, but not systematic uncertainty, which needs to be minimized by adequate modeling to obtain accurate credible intervals for MD observables. Our results represent a step towards accurate UQ that is of vital importance for trustworthy NN potential-based MD simulations required for decision-making in practice.

physics.chem-ph↗

arXiv2022

Deep Coarse-grained Potentials via Relative Entropy Minimization

Neural network (NN) potentials are a natural choice for coarse-grained (CG) models. Their many-body capacity allows highly accurate approximations of the potential of mean force, promising CG simulations at unprecedented accuracy. CG NN potentials trained bottom-up via force matching (FM), however, suffer from finite data effects: They rely on prior potentials for physically sound predictions outside the training data domain and the corresponding free energy surface is sensitive to errors in transition regions. The standard alternative to FM for classical potentials is relative entropy (RE) minimization, which has not yet been applied to NN potentials. In this work, we demonstrate for benchmark problems of liquid water and alanine dipeptide that RE training is more data efficient due to accessing the CG distribution during training, resulting in improved free energy surfaces and reduced sensitivity to prior potentials. In addition, RE learns to correct time integration errors, allowing larger time steps in CG molecular dynamics simulation while maintaining accuracy. Thus, our findings support the use of training objectives beyond FM as a promising direction for improving CG NN potential accuracy and reliability.

physics.chem-ph↗

arXiv2021

Learning neural network potentials from experimental data via Differentiable Trajectory Reweighting

In molecular dynamics (MD), neural network (NN) potentials trained bottom-up on quantum mechanical data have seen tremendous success recently. Top-down approaches that learn NN potentials directly from experimental data have received less attention, typically facing numerical and computational challenges when backpropagating through MD simulations. We present the Differentiable Trajectory Reweighting (DiffTRe) method, which bypasses differentiation through the MD simulation for time-independent observables. Leveraging thermodynamic perturbation theory, we avoid exploding gradients and achieve around 2 orders of magnitude speed-up in gradient computation for top-down learning. We show effectiveness of DiffTRe in learning NN potentials for an atomistic model of diamond and a coarse-grained model of water based on diverse experimental observables including thermodynamic, structural and mechanical properties. Importantly, DiffTRe also generalizes bottom-up structural coarse-graining methods such as iterative Boltzmann inversion to arbitrary potentials. The presented method constitutes an important milestone towards enriching NN potentials with experimental data, particularly when accurate bottom-up data is unavailable.

physics.chem-ph↗

arXiv2019

Sparse Identification of Truncation Errors

This work presents a data-driven approach to the identification of spatial and temporal truncation errors for linear and nonlinear discretization schemes of Partial Differential Equations (PDEs). Motivated by the central role of truncation errors, for example in the creation of implicit Large Eddy schemes, we introduce the Sparse Identification of Truncation Errors (SITE) framework to automatically identify the terms of the modified differential equation from simulation data. We build on recent advances in the field of data-driven discovery and control of complex systems and combine it with classical work on modified differential equation analysis of Warming, Hyett, Lerat and Peyret. We augment a sparse regression-rooted approach with appropriate preconditioning routines to aid in the identification of the individual modified differential equation terms. The construction of such a custom algorithm pipeline allows attenuating of multicollinearity effects as well as automatic tuning of the sparse regression hyperparameters using the Bayesian information criterion (BIC). As proof of concept, we constrain the analysis to finite difference schemes and leave other numerical schemes open for future inquiry. Test cases include the linear advection equation with a forward-time, backward-space discretization, the Burgers' equation with a MacCormack predictor-corrector scheme and the Korteweg-de Vries equation with a Zabusky and Kruska discretization scheme. Based on variation studies, we derive guidelines for the selection of discretization parameters, preconditioning approaches and sparse regression algorithms. The results showcase highly accurate predictions underlining the promise of SITE for the analysis and optimization of discretization schemes, where analytic derivation of modified differential equations is infeasible.

math.NA↗