We introduce M$^2$Hub, a toolkit for advancing machine learning in materials discovery. Machine learning has achieved remarkable progress in modeling molecular structures, especially biomolecules for drug discovery. However, the development of machine learning approaches for modeling materials structures lag behind, which is partly due to the lack of an integrated platform that enables access to diverse tasks for materials discovery. To bridge this gap, M$^2$Hub will enable easy access to materials discovery tasks, datasets, machine learning methods, evaluations, and benchmark results that cover the entire workflow. Specifically, the first release of M$^2$Hub focuses on three key stages in materials discovery: virtual screening, inverse design, and molecular simulation, including 9 datasets that covers 6 types of materials with 56 tasks across 8 types of material properties. We further provide 2 synthetic datasets for the purpose of generative tasks on materials. In addition to random data splits, we also provide 3 additional data partitions to reflect the real-world materials discovery scenarios. State-of-the-art machine learning methods (including those are suitable for materials structures but never compared in the literature) are benchmarked on representative tasks. Our codes and library are publicly available at https://github.com/yuanqidu/M2Hub.
Artificial intelligence for scientific discovery has recently generated significant interest within the machine learning and scientific communities, particularly in the domains of chemistry, biology, and material discovery. For these scientific problems, molecules serve as the fundamental building blocks, and machine learning has emerged as a highly effective and powerful tool for modeling their geometric structures. Nevertheless, due to the rapidly evolving process of the field and the knowledge gap between science (e.g., physics, chemistry, & biology) and machine learning communities, a benchmarking study on geometrical representation for such data has not been conducted. To address such an issue, in this paper, we first provide a unified view of the current symmetry-informed geometric methods, classifying them into three main categories: invariance, equivariance with spherical frame basis, and equivariance with vector frame basis. Then we propose a platform, coined Geom3D, which enables benchmarking the effectiveness of geometric strategies. Geom3D contains 16 advanced symmetry-informed geometric representation models and 14 geometric pretraining methods over 46 diverse datasets, including small molecules, proteins, and crystalline materials. We hope that Geom3D can, on the one hand, eliminate barriers for machine learning researchers interested in exploring scientific problems; and, on the other hand, provide valuable guidance for researchers in computational chemistry, structural biology, and materials science, aiding in the informed selection of representation techniques for specific applications.
Accelerated discovery with machine learning (ML) has begun to provide the advances in efficiency needed to overcome the combinatorial challenge of computational materials design. Nevertheless, ML-accelerated discovery both inherits the biases of training data derived from density functional theory (DFT) and leads to many attempted calculations that are doomed to fail. Many compelling functional materials and catalytic processes involve strained chemical bonds, open-shell radicals and diradicals, or metal-organic bonds to open-shell transition-metal centers. Although promising targets, these materials present unique challenges for electronic structure methods and combinatorial challenges for their discovery. In this Perspective, we describe the advances needed in accuracy, efficiency, and approach beyond what is typical in conventional DFT-based ML workflows. These challenges have begun to be addressed through ML models trained to predict the results of multiple methods or the differences between them, enabling quantitative sensitivity analysis. For DFT to be trusted for a given data point in a high-throughput screen, it must pass a series of tests. ML models that predict the likelihood of calculation success and detect the presence of strong correlation will enable rapid diagnoses and adaptation strategies. These "decision engines" represent the first steps toward autonomous workflows that avoid the need for expert determination of the robustness of DFT-based materials discoveries.
Transition state (TS) search is key in chemistry for elucidating reaction mechanisms and exploring reaction networks. The search for accurate 3D TS structures, however, requires numerous computationally intensive quantum chemistry calculations due to the complexity of potential energy surfaces. Here, we developed an object-aware SE(3) equivariant diffusion model that satisfies all physical symmetries and constraints for generating sets of structures - reactant, TS, and product - in an elementary reaction. Provided reactant and product, this model generates a TS structure in seconds instead of hours required when performing quantum chemistry-based optimizations. The generated TS structures achieve a median of 0.08 Å root mean square deviation compared to the true TS. With a confidence scoring model for uncertainty quantification, we approach an accuracy required for reaction rate estimation (2.6 kcal/mol) by only performing quantum chemistry-based optimizations on 14\% of the most challenging reactions. We envision the proposed approach useful in constructing large reaction networks with unknown mechanisms.
High-throughput screening of large hypothetical databases of metal-organic frameworks (MOFs) can uncover new materials, but their stability in real-world applications is often unknown. We leverage community knowledge and machine learning (ML) models to identify MOFs that are thermally stable and stable upon activation. We separate these MOFs into their building blocks and recombine them to make a new hypothetical MOF database of over 50,000 structures that samples orders of magnitude more connectivity nets and inorganic building blocks than prior databases. This database shows an order of magnitude enrichment of ultrastable MOF structures that are stable upon activation and more than one standard deviation more thermally stable than the average experimentally characterized MOF. For the nearly 10,000 ultrastable MOFs, we compute bulk elastic moduli to confirm these materials have good mechanical stability, and we report methane deliverable capacities. Our work identifies privileged metal nodes in ultrastable MOFs that optimize gas storage and mechanical stability simultaneously.
Virtual high throughput screening (VHTS) and machine learning (ML) have greatly accelerated the design of single-site transition-metal catalysts. VHTS of catalysts, however, is often accompanied with high calculation failure rate and wasted computational resources due to the difficulty of simultaneously converging all mechanistically relevant reactive intermediates to expected geometries and electronic states. We demonstrate a dynamic classifier approach, i.e., a convolutional neural network that monitors geometry optimization on the fly, and exploit its good performance and transferability for catalyst design. We show that the dynamic classifier performs well on all reactive intermediates in the representative catalytic cycle of the radical rebound mechanism for methane-to-methanol despite being trained on only one reactive intermediate. The dynamic classifier also generalizes to chemically distinct intermediates and metal centers absent from the training data without loss of accuracy or model confidence. We rationalize this superior model transferability to the use of on-the-fly electronic structure and geometric information generated from density functional theory calculations and the convolutional layer in the dynamic classifier. Combined with model uncertainty quantification, the dynamic classifier saves more than half of the computational resources that would have been wasted on unsuccessful calculations for all reactive intermediates being considered.
Chenru Duan, Aditya Nandy, Gianmarco Terrones et al.
Transition metal chromophores with earth-abundant transition metals are an important design target for their applications in lighting and non-toxic bioimaging, but their design is challenged by the scarcity of complexes that simultaneously have optimal target absorption energies in the visible region as well as well-defined ground states. Machine learning (ML) accelerated discovery could overcome such challenges by enabling screening of a larger space, but is limited by the fidelity of the data used in ML model training, which is typically from a single approximate density functional. To address this limitation, we search for consensus in predictions among 23 density functional approximations across multiple rungs of Jacobs ladder. To accelerate the discovery of complexes with absorption energies in the visible region while minimizing MR character, we use 2D efficient global optimization to sample candidate low-spin chromophores from multi-million complex spaces. Despite the scarcity (i.e., approx. 0.01\%) of potential chromophores in this large chemical space, we identify candidates with high likelihood (i.e., > 10\%) of computational validation as the ML models improve during active learning, representing a 1,000-fold acceleration in discovery. Absorption spectra of promising chromophores from time-dependent density functional theory verify that 2/3 of candidates have the desired excited state properties. The observation that constituent ligands from our leads have demonstrated interesting optical properties in the literature exemplifies the effectiveness of our construction of a realistic design space and active learning approach.
Approximate density functional theory (DFT) has become indispensable owing to its cost-accuracy trade-off in comparison to more computationally demanding but accurate correlated wavefunction theory. To date, however, no single density functional approximation (DFA) with universal accuracy has been identified, leading to uncertainty in the quality of data generated from DFT. With electron density fitting and transfer learning, we build a DFA recommender that selects the DFA with the lowest expected error with respect to gold standard but cost-prohibitive coupled cluster theory in a system-specific manner. We demonstrate this recommender approach on vertical spin-splitting energy evaluation for challenging transition metal complexes. Our recommender predicts top-performing DFAs and yields excellent accuracy (ca. 2 kcal/mol) for chemical discovery, outperforming both individual transfer learning models and the single best functional in a set of 48 DFAs. We demonstrate the transferability of the DFA recommender to experimentally synthesized compounds with distinct chemistry.
Chenru Duan, Adriana J. Ladera, Julian C. -L. Liu et al.
Accurate virtual high-throughput screening (VHTS) of transition metal complexes (TMCs) remains challenging due to the possibility of high multi-reference (MR) character that complicates property evaluation. We compute MR diagnostics for over 5,000 ligands present in previously synthesized transition metal complexes in the Cambridge Structural Database (CSD). To accomplish this task, we introduce an iterative approach for consistent ligand charge assignment for ligands in the CSD. Across this set, we observe that MR character correlates linearly with the inverse value of the averaged bond order over all bonds in the molecule. We then demonstrate that ligand additivity of MR character holds in TMCs, which suggests that the TMC MR character can be inferred from the sum of the MR character of the ligands. Encouraged by this observation, we leverage ligand additivity and develop a ligand-derived machine learning representation to train neural networks to predict the MR character of TMCs from properties of the constituent ligands. This approach yields models with excellent performance and superior transferability to unseen ligand chemistry and compositions.
Stephen G. Dale, Nikita Kazeev, Alastair J. A. Price et al.
Artificial intelligence and machine learning are reshaping how we approach scientific discovery, not by replacing established methods but by extending what researchers can probe, predict, and design. In this roadmap we provide a forward-looking view of AI-enabled science across biology, chemistry, climate science, mathematics, materials science, physics, self-driving laboratories and unconventional computing. Several shared themes emerge: the need for diverse and trustworthy data, transferable electronic-structure and interatomic models, AI systems integrated into end-to-end scientific workflows that connect simulations to experiments and generative systems grounded in synthesisability rather than purely idealised phases. Across domains, we highlight how large foundation models, active learning and self-driving laboratories can close loops between prediction and validation while maintaining reproducibility and physical interpretability. Taken together, these perspectives outline where AI-enabled science stands today, identify bottlenecks in data, methods and infrastructure, and chart concrete directions for building AI systems that are not only more powerful but also more transparent and capable of accelerating discovery in complex real-world environments.
Gianmarco Terrones, Chenru Duan, Aditya Nandy et al.
Photoactive iridium complexes are of broad interest due to their applications ranging from lighting to photocatalysis. However, the excited state property prediction of these complexes challenges ab initio methods such as time-dependent density functional theory (TDDFT) both from an accuracy and a computational cost perspective, complicating high throughput virtual screening (HTVS). We instead leverage low-cost machine learning (ML) models to predict the excited state properties of photoactive iridium complexes. We use experimental data of 1,380 iridium complexes to train and evaluate the ML models and identify the best-performing and most transferable models to be those trained on electronic structure features from low-cost density functional theory tight binding calculations. Using these models, we predict the three excited state properties considered, mean emission energy of phosphorescence, excited state lifetime, and emission spectral integral, with accuracy competitive with or superseding TDDFT. We conduct feature importance analysis to identify which iridium complex attributes govern excited state properties and we validate these trends with explicit examples. As a demonstration of how our ML models can be used for HTVS and the acceleration of chemical discovery, we curate a set of novel hypothetical iridium complexes and identify promising ligands for the design of new phosphors.
Recent progress of deep generative models in the vision and language domain has stimulated significant interest in more structured data generation such as molecules. However, beyond generating new random molecules, efficient exploration and a comprehensive understanding of the vast chemical space are of great importance to molecular science and applications in drug design and materials discovery. In this paper, we propose a new framework, ChemFlow, to traverse chemical space through navigating the latent space learned by molecule generative models through flows. We introduce a dynamical system perspective that formulates the problem as learning a vector field that transports the mass of the molecular distribution to the region with desired molecular properties or structure diversity. Under this framework, we unify previous approaches on molecule latent space traversal and optimization and propose alternative competing methods incorporating different physical priors. We validate the efficacy of ChemFlow on molecule manipulation and single- and multi-objective molecule optimization tasks under both supervised and unsupervised molecular discovery settings. Codes and demos are publicly available on GitHub at https://github.com/garywei944/ChemFlow.
Molecular discovery, when formulated as an optimization problem, presents significant computational challenges because optimization objectives can be non-differentiable. Evolutionary Algorithms (EAs), often used to optimize black-box objectives in molecular discovery, traverse chemical space by performing random mutations and crossovers, leading to a large number of expensive objective evaluations. In this work, we ameliorate this shortcoming by incorporating chemistry-aware Large Language Models (LLMs) into EAs. Namely, we redesign crossover and mutation operations in EAs using LLMs trained on large corpora of chemical information. We perform extensive empirical studies on both commercial and open-source models on multiple tasks involving property optimization, molecular rediscovery, and structure-based drug design, demonstrating that the joint usage of LLMs with EAs yields superior performance over all baseline models across single- and multi-objective settings. We demonstrate that our algorithm improves both the quality of the final solution and convergence speed, thereby reducing the number of required objective evaluations. Our code is available at http://github.com/zoom-wang112358/MOLLEO
Crystal structure generation is fundamental to materials science, enabling the discovery of novel materials with desired properties. While existing approaches leverage Large Language Models (LLMs) through extensive fine-tuning on materials databases, we show that pre-trained LLMs can inherently generate novel and stable crystal structures without additional fine-tuning. Our framework employs LLMs as intelligent proposal agents within an evolutionary pipeline that guides them to perform implicit crossover and mutation operations while maintaining chemical validity. We demonstrate that MatLLMSearch achieves a 78.38% metastable rate validated by machine learning interatomic potentials and 31.7% DFT-verified stability, outperforming specialized models such as CrystalTextLLM. Beyond crystal structure generation, we further demonstrate that our framework adapts to diverse materials design tasks, including crystal structure prediction and multi-objective optimization of properties such as deformation energy and bulk modulus, all without fine-tuning. These results establish our framework as a versatile and effective framework for consistent high-quality materials discovery, offering training-free generation of novel stable structures with reduced overhead and broader accessibility.
Designing functional transition metal complexes (TMCs) faces challenges due to the vast search space of metals and ligands, requiring efficient optimization strategies. Traditional genetic algorithms (GAs) are commonly used, employing random mutations and crossovers driven by explicit mathematical objectives to explore this space. Transferring knowledge between different GA tasks, however, is difficult. We integrate large language models (LLMs) into the evolutionary optimization framework (LLM-EO) and apply it in both single- and multi-objective optimization for TMCs. We find that LLM-EO surpasses traditional GAs by leveraging the chemical knowledge of LLMs gained during their extensive pretraining. Remarkably, without supervised fine-tuning, LLMs utilize the full historical data from optimization processes, outperforming those focusing only on top-performing TMCs. LLM-EO successfully identifies eight of the top-20 TMCs with the largest HOMO-LUMO gaps by proposing only 200 candidates out of a 1.37 million TMCs space. Through prompt engineering using natural language, LLM-EO introduces unparalleled flexibility into multi-objective optimizations, thereby circumventing the necessity for intricate mathematical formulations. As generative models, LLMs can suggest new ligands and TMCs with unique properties by merging both internal knowledge and external chemistry data, thus combining the benefits of efficient optimization and molecular generation. With increasing potential of LLMs as pretrained foundational models and new post-training inference strategies, we foresee broad applications of LLM-based evolutionary optimization in chemistry and materials design.
Transition states (TSs) are transient structures that are key in understanding reaction mechanisms and designing catalysts but challenging to be captured in experiments. Alternatively, many optimization algorithms have been developed to search for TSs computationally. Yet the cost of these algorithms driven by quantum chemistry methods (usually density functional theory) is still high, posing challenges for their applications in building large reaction networks for reaction exploration. Here we developed React-OT, an optimal transport approach for generating unique TS structures from reactants and products. React-OT generates highly accurate TS structures with a median structural root mean square deviation (RMSD) of 0.053Å and median barrier height error of 1.06 kcal/mol requiring only 0.4 second per reaction. The RMSD and barrier height error is further improved by roughly 25\% through pretraining React-OT on a large reaction dataset obtained with a lower level of theory, GFN2-xTB. We envision that the remarkable accuracy and rapid inference of React-OT will be highly useful when integrated with the current high-throughput TS search workflow. This integration will facilitate the exploration of chemical reactions with unknown mechanisms.
Advances in generative artificial intelligence are transforming how metal-organic frameworks (MOFs) are designed and discovered. This Perspective introduces the shift from laborious enumeration of MOF candidates to generative approaches that can autonomously propose and synthesize in the laboratory new porous reticular structures on demand. We outline the progress of employing deep learning models, such as variational autoencoders, diffusion models, and large language model-based agents, that are fueled by the growing amount of available data from the MOF community and suggest novel crystalline materials designs. These generative tools can be combined with high-throughput computational screening and even automated experiments to form accelerated, closed-loop discovery pipelines. The result is a new paradigm for reticular chemistry in which AI algorithms more efficiently direct the search for high-performance MOF materials for clean air and energy applications. Finally, we highlight remaining challenges such as synthetic feasibility, dataset diversity, and the need for further integration of domain knowledge.
Chenru Duan, Daniel B. K. Chu, Aditya Nandy et al.
Appropriately identifying and treating molecules and materials with significant multi-reference (MR) character is crucial for achieving high data fidelity in virtual high throughput screening (VHTS). Nevertheless, most VHTS is carried out with approximate density functional theory (DFT) using a single functional. Despite development of numerous MR diagnostics, the extent to which a single value of such a diagnostic indicates MR effect on chemical property prediction is not well established. We evaluate MR diagnostics of over 10,000 transition metal complexes (TMCs) and compare to those in organic molecules. We reveal that only some MR diagnostics are transferable across these materials spaces. By studying the influence of MR character on chemical properties (i.e., MR effect) that involves multiple potential energy surfaces (i.e., adiabatic spin splitting, $ΔE_\mathrm{H-L}$, and ionization potential, IP), we observe that cancellation in MR effect outweighs accumulation. Differences in MR character are more important than the total degree of MR character in predicting MR effect in property prediction. Motivated by this observation, we build transfer learning models to directly predict CCSD(T)-level adiabatic $ΔE_\mathrm{H-L}$ and IP from lower levels of theory. By combining these models with uncertainty quantification and multi-level modeling, we introduce a multi-pronged strategy that accelerates data acquisition by at least a factor of three while achieving chemical accuracy (i.e., 1 kcal/mol) for robust VHTS.
Machine learning (ML)-accelerated discovery requires large amounts of high-fidelity data to reveal predictive structure-property relationships. For many properties of interest in materials discovery, the challenging nature and high cost of data generation has resulted in a data landscape that is both scarcely populated and of dubious quality. Data-driven techniques starting to overcome these limitations include the use of consensus across functionals in density functional theory, the development of new functionals or accelerated electronic structure theories, and the detection of where computationally demanding methods are most necessary. When properties cannot be reliably simulated, large experimental data sets can be used to train ML models. In the absence of manual curation, increasingly sophisticated natural language processing and automated image analysis are making it possible to learn structure-property relationships from the literature. Models trained on these data sets will improve as they incorporate community feedback.
Although the tailored metal active sites and porous architectures of MOFs hold great promise for engineering challenges ranging from gas separations to catalysis, a lack of understanding of how to improve their stability limits their use in practice. To overcome this limitation, we extract thousands of published reports of the key aspects of MOF stability necessary for their practical application: the ability to withstand high temperatures without degrading and the capacity to be activated by removal of solvent molecules. From nearly 4,000 manuscripts, we use natural language processing and automated image analysis to obtain over 2,000 solvent-removal stability measures and 3,000 thermal degradation temperatures. We analyze the relationships between stability properties and the chemical and geometric structures in this set to identify limits of prior heuristics derived from smaller sets of MOFs. By training predictive machine learning (ML, i.e., Gaussian process and artificial neural network) models to encode the structure-property relationships with graph- and pore-structure-based representations, we are able to make predictions of stability orders of magnitude faster than conventional physics-based modeling or experiment. Interpretation of important features in ML models provides insights that we use to identify strategies to engineer increased stability into typically unstable 3d-containing MOFs that are frequently targeted for catalytic applications. We expect our approach to accelerate the time to discovery of stable, practical MOF materials for a wide range of applications.
Chenru Duan, Shuxin Chen, Michael G. Taylor et al.
Computational virtual high-throughput screening (VHTS) with density functional theory (DFT) and machine-learning (ML)-acceleration is essential in rapid materials discovery. By necessity, efficient DFT-based workflows are carried out with a single density functional approximation (DFA). Nevertheless, properties evaluated with different DFAs can be expected to disagree for the cases with challenging electronic structure (e.g., open shell transition metal complexes, TMCs) for which rapid screening is most needed and accurate benchmarks are often unavailable. To quantify the effect of DFA bias, we introduce an approach to rapidly obtain property predictions from 23 representative DFAs spanning multiple families and "rungs" (e.g., semi-local to double hybrid) and basis sets on over 2,000 TMCs. Although computed properties (e.g., spin-state ordering and frontier orbital gap) naturally differ by DFA, high linear correlations persist across all DFAs. We train independent ML models for each DFA and observe convergent trends in feature importance; these features thus provide DFA-invariant, universal design rules. We devise a strategy to train ML models informed by all 23 DFAs and use them to predict properties (e.g., spin-splitting energy) of over 182k TMCs. By requiring consensus of the ANN-predicted DFA properties, we improve correspondence of these computational lead compounds with literature-mined, experimental compounds over the single-DFA approach typically employed. Both feature analysis and consensus-based ML provide efficient, alternative paths to overcome accuracy limitations of practical DFT.
Daniel R. Harper, Aditya Nandy, Naveen Arunachalam et al.
Strategies for machine-learning(ML)-accelerated discovery that are general across materials composition spaces are essential, but demonstrations of ML have been primarily limited to narrow composition variations. By addressing the scarcity of data in promising regions of chemical space for challenging targets like open-shell transition-metal complexes, general representations and transferable ML models that leverage known relationships in existing data will accelerate discovery. Over a large set (ca. 1000) of isovalent transition-metal complexes, we quantify evident relationships for different properties (i.e., spin-splitting and ligand dissociation) between rows of the periodic table (i.e., 3d/4d metals and 2p/3p ligands). We demonstrate an extension to graph-based revised autocorrelation (RAC) representation (i.e., eRAC) that incorporates the effective nuclear charge alongside the nuclear charge heuristic that otherwise overestimates dissimilarity of isovalent complexes. To address the common challenge of discovery in a new space where data is limited, we introduce a transfer learning approach in which we seed models trained on a large amount of data from one row of the periodic table with a small number of data points from the additional row. We demonstrate the synergistic value of the eRACs alongside this transfer learning strategy to consistently improve model performance. Analysis of these models highlights how the approach succeeds by reordering the distances between complexes to be more consistent with the periodic table, a property we expect to be broadly useful for other materials domains.