arXiv subjects

Teresa Head-Gordon

Publications and source records attributed to Teresa Head-Gordon.

At least 19 recordsLinked to original sources

arXiv2026

How Well Can Frontier Large Language Models Generate Structures? High Quality Prediction of Molecular Geometries with Help from Fine-Tuning

The power of Large Language Models (LLMs) has led us to investigate how they might be fine-tuned for learning the "language of molecular geometry". The fine-tuning of the LLMs using Cartesian coordinates or Z-matrices provides an extremely simple method for accurately predicting equilibrium structures and diverse sets of conformers of small organic and drug-like molecules with excellent accuracy and outperforming specialized deep learning models. While the most common representation of molecular geometries are Cartesian coordinates performs adequately, we find that the inherent invariances and relational nature of geometries represented as Z-matrices provides a better grammar for LLM adaptation. Finally, we show that enhancing an LLMs capabilities for robust prediction of small molecule geometries still retains nearly all of its pre-trained language abilities by randomly mixing in small quantities of natural language prompt-response pairs into the fine-tuning.

physics.chem-ph↗

arXiv2026

Towards Diverse Scientific Hypothesis Search with Large Language Models

Large language models (LLMs) are on the rise for accelerating scientific discovery, most recently in advanced tasks such as generating valid scientific hypotheses. Yet in many discovery settings, the goal is not to identify a single best hypothesis since validation can be noisy and expensive, and scientists benefit from a set of high-quality alternative hypotheses that hedge against downstream uncertainty for the best solutions. Nevertheless, commonly used evolutionary search recipes tend to prioritize optimization over exploration in hypothesis generation, and the resulting selection pressure during the search process leads to diversity collapse. Motivated by these limitations, we formulate hypothesis search as a sampling problem, where the objective is to efficiently produce diverse, high-quality hypotheses under a fixed validation budget. Building on this perspective, we propose \ours, an evolutionary framework inspired by the classical parallel tempering algorithm that searches hypotheses at multiple temperature levels and enables principled information exchange across temperatures to improve exploration without disrupting convergence. Across domains including molecular discovery, equation discovery, and algorithm discovery, our approach consistently improves both hypothesis quality and diversity under the same validation budget, and produces candidates that remain robust under more expensive downstream computational validations.

cs.LG↗

arXiv2026

Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.

cs.AI↗

arXiv2026

Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding

The reliable determination of transition states (TSs) benefits from second-order information for robust convergence and validation, but the computational expense of Hessians prohibits their routine use in TS optimization. Here, we present a machine-learning-driven TS optimizer that directly predicts the leftmost Hessian eigenvector (LMHE), the critical mode that locally approximates the reaction coordinate encompassing the TS. We demonstrate that our LMHE optimizer recovers TS solutions at the same rate as full Hessian optimizers, and more robustly from degraded initial guess geometries, thereby eliminating the excessively long wall times characteristic of full-Hessian approaches and reducing total gradient evaluations compared to standard quasi-Newton methods. We further improve accuracy and robustness using uncertainty quantification for identifying occasional LMHE prediction failures, that then falls back to a full Hessian update from the machine learned potential at that optimization step, avoiding expensive active learning. Overall our methodology and semi-automated workflow delivers second-order stability at first-order computational expense to provide a highly efficient engine for high-throughput reaction discovery.

physics.chem-ph↗

arXiv2026

DrugPlayGround: Benchmarking Large Language Models and Embeddings for Drug Discovery

Large language models (LLMs) are in the ascendancy for research in drug discovery, offering unprecedented opportunities to reshape drug research by accelerating hypothesis generation, optimizing candidate prioritization, and enabling more scalable and cost-effective drug discovery pipelines. However there is currently a lack of objective assessments of LLM performance to ascertain their advantages and limitations over traditional drug discovery platforms. To tackle this emergent problem, we have developed DrugPlayGround, a framework to evaluate and benchmark LLM performance for generating meaningful text-based descriptions of physiochemical drug characteristics, drug synergism, drug-protein interactions, and the physiological response to perturbations introduced by drug molecules. Moreover, DrugPlayGround is designed to work with domain experts to provide detailed explanations for justifying the predictions of LLMs, thereby testing LLMs for chemical and biological reasoning capabilities to push their greater use at the frontier of drug discovery at all of its stages.

cs.LG↗

arXiv2026

The Interplay of Pauli Repulsion, Electrostatics, and Field Inhomogeneity for Blueshifting and Redshifting Vibrational Probe Molecules

Many molecules' vibrational frequencies are sensitive to intermolecular electric fields, enabling them to probe the field in complex molecular environments. However, it is often unclear whether the probe is responding to the local electric field or other types of intermolecular interactions, inhibiting interpretation of the frequency and effectiveness as probes. This is especially true of molecules whose vibrational frequencies blueshift instead of the more typical redshift in hydrogen bonding configurations. Here we computationally investigate the causes of redshifting versus blueshifting over a range of vibrational reporters. First, we apply adiabatic energy decomposition analysis to a paradigmatic set of probes, finding that redshifting only occurs when electrostatic interactions are strong enough to overcome the dominant and large blueshifting contribution of Pauli repulsion. Furthermore, we demonstrate that field inhomogeneity can further shift the frequency of many probes substantially to either reinforce or counteract the shift expected from a homogeneous field. We find that redshifting is reinforced by electric field inhomogeneity, otherwise field inhomogeneity further weakens the electrostatic contribution relative to Pauli repulsion, leading to blueshifting. Further calculations indicate that the probe's response to field inhomogeneity can be understood by considering the mass of the atoms involved in the stretching mode and sign of the electric field. In explaining the interplay of different intermolecular interactions and field inhomogeneity for many probes, our results should enable the use and interpretation of spectroscopic probes and their connection to electric fields in more complex systems.

physics.chem-ph↗

arXiv2025

Accelerating Scientific Discovery with Autonomous Goal-evolving Agents

There has been unprecedented interest in developing agents that expand the boundary of scientific discovery, primarily by optimizing quantitative objective functions specified by scientists. However, for grand challenges in science, these objectives may only be imperfect proxies. We argue that automating objective function design is a central, yet unmet need for scientific discovery agents. In this work, we introduce the Scientific Autonomous Goal-evolving Agent (SAGA) to address this challenge. SAGA employs a bi-level architecture in which an outer loop of LLM agents analyzes optimization outcomes, proposes new objectives, and converts them into computable scoring functions, while an inner loop performs solution optimization under the current objectives. This bi-level design enables systematic exploration of the space of objectives and their trade-offs, rather than treating them as fixed inputs. We demonstrate the framework through a wide range of design applications, including antibiotics, nanobodies, functional DNA sequences, inorganic materials, and chemical processes. Notably, our experimental validation identifies a structurally novel hit with promising potency and safety profiles for E. coli in the antibiotic design task, and three de novo PD-L1 binders in the nanobody design task. These results suggest that automating objective formulation can substantially improve the effectiveness of scientific discovery agents.

cs.AI↗

arXiv2025

Teachers that teach the irrelevant: Pre-training machine learned interaction potentials with classical force fields for robust molecular dynamics simulations

Machine learned interaction potentials (MLIPs) have become a critical component of large-scale, high-quality simulations for a range of chemical and biochemical systems. Yet, despite their in-distribution accuracy, molecular dynamics simulations using MLIPs exhibit numerical instabilities due to underlying data insufficiencies when encountering new regions of the potential energy surface. Here we propose a pre-training learning scheme that uses low-quality, practically free, single-molecule non-reactive force field data while all intermolecular interactions and reactive properties are learned at a fine-tuning stage with a small amount of computationally more expensive labels. We show that the force field pre-training approach followed by data efficient ab initio fine tuning allows for stable and accurate molecular dynamics and metadynamics simulations of gas phase molecules, liquid water, and hydrogen combustion reactions compared to models trained from scratch.

physics.chem-ph↗

arXiv2025

Improved Treatment of 1-4 interactions in Force Fields for Molecular Dynamics Simulations

Traditional force fields commonly use a combination of bonded torsional terms and empirically scaled non-bonded interactions to capture 1-4 energies and forces of atoms separated by three bonds in a molecule. While this approach can yield accurate torsional energy barriers, it often leads to inaccurate forces and erroneous geometries, and creates an interdependence between dihedral terms and non-bonded interactions, complicating parameterization and reducing transferability. In this paper, we demonstrate that 1-4 interactions can be accurately modeled using only bonded coupling terms, eliminating the need for arbitrarily scaled non-bonded interactions altogether. Furthermore by leveraging the automated parameterization capabilities of the Q-Force toolkit, we efficiently determine the necessary coupling terms without the need for manual adjustment. Our approach is first validated on a range of small molecule systems, encompassing both flexible and rigid structures, and shows a significant improvement in force field accuracy, obtaining sub-kcal/mol mean absolute error for every molecule tested. We further extend the bonded-only model for 1-4 interactions to Amber ff14sb, CHARMM36, and OPLS-AA force fields to reproduce ab initio gas and implicit solvent $\phi,\psi$ surfaces of alanine dipeptide.

physics.chem-ph↗

arXiv2025

Can Charge Transfer Across C-H...O Hydrogen Bonds Stabilize Oil Droplets in Water?

Oil-water emulsions resist aggregation due to the presence of negative charges at their surface that leads to mutual repulsion between droplets, but the molecular origin of oil charge is currently under debate. While much evidence has suggested that ionic species must accumulate at the interface, an alternative perspective attributes the negative charge on the oil droplet to charge transfer of electron density from water to oil molecules. While the charge transfer mechanism is consistent with the correct sign of oil charge, it is just as important to provide good estimates of the charge magnitude to explain emulsion stability and electrophoresis experiments. Here we show using energy decomposition analysis that the amount of net flow of charge from water to oil is negligibly small due to nearly equal forward and backward charge transfer through weak oil-water interactions, such that oil droplets would be unstable and coalesce, contrary to experiment. The lack of charge transfer also explains why vibrational sum frequency scattering reports a blue shift in the oil C-H frequency when forming emulsions with water, which arises from Pauli repulsion due to localized confinement at the interface. Finally, unlike ions, neither charge transfer nor dynamic polarization can produce a finite conductivity needed to couple to electric fields that would explain electrophoretic mobility.

physics.chem-ph↗

arXiv2025

Ammonia Synthesis under Ambient Conditions: Insights into Water-Nitrogen-Magnetite Interfaces

New routes for transforming nitrogen into ammonia at ambient conditions would be a milestone toward an energy efficient and economically attractive production route in comparison to the traditional Haber-Bosch process. Recently, the synthesis of ammonia from water and nitrogen at room temperature and atmospheric pressure has been reported to be catalyzed by Fe3O4 at the air-water interface. By integrating ambient pressure X-ray photoelectron spectroscopy and ab initio molecular dynamics and free energy calculations, we investigate the underlying thermodynamic mechanisms governing ammonia and hydrazine formation at the water-Fe3O4-nanoparticle interface. We find that, unlike pure Fe3O4 where N2 can only interact with a limited number of Fe sites, hydroxylated species introduce large and diverse adsorption geometries where N2 can bind through either Fe sites or Fe-OH groups, each of which are capable of independently facilitating proton-coupled electron transfer.

physics.chem-ph↗

arXiv2025

SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models

Generative machine learning models for exploring chemical space have shown immense promise, but many molecules they generate are too difficult to synthesize, making them impractical for further investigation or development. In this work, we present a novel approach by fine-tuning Meta's Llama3 Large Language Models (LLMs) to create SynLlama, which generates full synthetic pathways made of commonly accessible building blocks and robust organic reaction templates. SynLlama explores a large synthesizable space using significantly less data, and offers strong performance in both forward and bottom-up synthesis planning compared to other state-of-the-art methods. We find that SynLlama, even without training on external building blocks, can effectively generalize to unseen yet purchasable building blocks, meaning that its reconstruction capabilities extend to a broader synthesizable chemical space than the training data. We also demonstrate the use of SynLlama in a pharmaceutical context for synthesis planning of analog molecules and hit expansion leads for proposed inhibitors of target proteins, offering medicinal chemists a valuable tool for discovery.

cs.LG↗

arXiv2025

Foundation Models for Atomistic Simulation of Chemistry and Materials

Given the power of large language and large vision models, it is of profound and fundamental interest to ask if a foundational model based on data and parameter scaling laws and pre-training strategies is possible for learned simulations of chemistry and materials. The scaling of large and diverse datasets and highly expressive architectures for chemical and materials sciences should result in a foundation model that is more efficient and broadly transferable, robust to out-of-distribution challenges, and easily fine-tuned to a variety of downstream observables, when compared to specific training from scratch on targeted applications in atomistic simulation. In this Perspective we aim to cover the rapidly advancing field of machine learned interatomic potentials (MLIP), and to illustrate a path to create chemistry and materials MLIP foundation models at larger scale.

physics.chem-ph↗

arXiv2025

Deep Learning of Proteins with Local and Global Regions of Disorder

Although machine learning has transformed protein structure prediction of folded protein ground states with remarkable accuracy, intrinsically disordered proteins and regions (IDPs/IDRs) are defined by diverse and dynamical structural ensembles that are predicted with low confidence by algorithms such as AlphaFold. We present a new machine learning method, IDPForge (Intrinsically Disordered Protein, FOlded and disordered Region GEnerator), that exploits a transformer protein language diffusion model to create all-atom IDP ensembles and IDR disordered ensembles that maintains the folded domains. IDPForge does not require sequence-specific training, back transformations from coarse-grained representations, nor ensemble reweighting, as in general the created IDP/IDR conformational ensembles show good agreement with solution experimental data, and options for biasing with experimental restraints are provided if desired. We envision that IDPForge with these diverse capabilities will facilitate integrative and structural studies for proteins that contain intrinsic disorder.

q-bio.BM↗

arXiv2024

Biological Insights from Integrative Modeling of Intrinsically Disordered Protein Systems

Intrinsically disordered proteins and regions are increasingly appreciated for their abundance in the proteome and the many functional roles they play in the cell. In this short review, we describe a variety of approaches used to obtain biological insight from the structural ensembles of disordered proteins, regions, and complexes and the integrative biology challenges that arise from combining diverse experiments and computational models. Importantly, we highlight findings regarding structural and dynamic characterization of disordered regions involved in binding and phase separation, as well as drug targeting of disordered regions, using a broad framework of integrative modeling approaches.

physics.bio-ph↗

arXiv2024

Determining the Role of Electrostatics in the Making and Breaking of the Caprin1-ATP Nanocondensate

We employ a multiscale computational approach to investigate the condensation process of the C-terminal low-complexity region of the Caprin1 protein as a function of increasing ATP concentration for three states: the initial mixed state, nanocondensate formation, and the dissolution of the droplet as it reenters the mixed state. We show that upon condensation ATP assembles via pi-pi interactions, resulting in the formation of a large cluster of stacked ATP molecules stabilized by sodium counterions. The surface of the ATP assembly interacts with the arginine-rich regions of the Caprin1 protein, particularly with its N-terminus, to promote the complete phase-separated droplet on a lengthscale of tens of nanometers. In order to understand droplet stability, we analyze the near-surface electrostatic potential (NS-ESP) of Caprin1 and estimate the zeta potential of the Caprin1-ATP assemblies. We predict a positive NS-ESP at the Caprin1 surface for low ATP concentrations that defines the early mixed state, in excellent agreement with the NS-ESP obtained from NMR experiments using paramagnetic resonance enhancement. By contrast, the NS-ESP of Caprin1 at the surface of the nanocondensate at moderate levels of ATP is highly negative compared to the mixed state, and estimates of a large zeta potential outside the highly dense region of charge further explains the remarkable stability of this phase separated droplet assembly. As ATP concentrations rise further, the strong electrostatic forces needed for nanocondensate stability are replaced by weaker Caprin1-ATP interactions that drive the reentry into the mixed state that exhibits a much lower zeta potential.

physics.bio-ph↗

arXiv2024

The Role of Interfaces and Charge for Chemical Reactivity in Microdroplets

A wide variety of reactions are reported to be dramatically accelerated in aqueous microdroplets, making them a promising platform for environmentally clean chemical synthesis. However to fully utilize the microdroplets for accelerating chemical reactions requires a fundamental understanding of how microdroplet chemistry differs from that of a homogeneous phase. Here we provide our perspective on recent progress to this end both experimentally and theoretically. We begin by reviewing the many ways in which microdroplets can be prepared, creating water/hydrophobic interfaces which have been frequently implicated in microdroplet reactivity due to preferential surface adsorption of solutes, persistent electric fields, and their acidity or basicity. These features of the interface interplay with specific mechanisms proposed for microdroplet reactivity, including partial solvation and possible gas phase channels. We especially highlight the role of droplet charge, which appears key to understanding how certain reactions, like the formation of hydrogen peroxide and reduced transition metal complexes, are thermodynamically possible in microdroplets. Lastly, we emphasize opportunities for theoretical advances in the microdroplet field generally, and to suggest experiments which would greatly enhance our understanding of this fascinating and emerging subject.

physics.chem-ph↗

arXiv2024

A Workflow to Create a High-Quality Protein-Ligand Binding Dataset for Training, Validation, and Prediction Tasks

Development of scoring functions (SFs) used to predict protein-ligand binding energies requires high-quality 3D structures and binding assay data for training and testing their parameters. In this work, we show that one of the widely-used datasets, PDBbind, suffers from several common structural artifacts of both proteins and ligands, which may compromise the accuracy, reliability, and generalizability of the resulting SFs. Therefore, we have developed a series of algorithms organized in a semi-automated workflow, HiQBind-WF, that curates non-covalent protein-ligand datasets to fix these problems. We also used this workflow to create an independent data set, HiQBind, by matching binding free energies from various sources including BioLiP, Binding MOAD and BindingDB with co-crystalized ligand-protein complexes from the PDB. The resulting HiQBind workflow and dataset are designed to ensure reproducibility and to minimize human intervention, while also being open-source to foster transparency in the improvements made to this important resource for the biology and drug discovery communities.

physics.bio-ph↗