Bridging Machine Learning and Thermodynamics for Accurate pK a Prediction

Integrating scientific principles into machine learning models to enhance their predictive performance and generalizability is a central challenge in the development of AI for Science. Herein, we introduce Uni-pK a, a novel framework that successfully incorporates thermodynamic principles into machine learning modeling, achieving high-precision predictions of acid dissociation constants (pK a), a crucial task in the rational design of drugs and catalysts, as well as a modeling challenge in computational physical chemistry for small organic molecules. Uni-pK a utilizes a comprehensive free energy model to represent molecular protonation equilibria accurately. It features a structure enumerator that reconstructs molecular configurations from pK a data, coupled with a neural network that functions as a free energy predictor, ensuring high-throughput, data-driven prediction while preserving thermodynamic consistency. Employing a pretraining-finetuning strategy with both predicted and experimental pK a data, Uni-pK a not only achieves state-of-the-art accuracy in chemoinformatics but also shows comparable precision to quantum mechanics-based methods.

MACE: Automated Assessment of Stereochemistry of Transition Metal Complexes and Its Applications in Computational Catalysis

Computational chemistry pipelines typically commence with geometry generation, well-established for organic compounds but presenting a considerable challenge for transition metal complexes. This paper introduces MACE, an automated computational workflow for converting chemist SMILES/MOL representations of the ligands and the metal center to 3D coordinates for all feasible stereochemical configurations for mononuclear octahedral and square planar complexes directly suitable for quantum chemical computations and implementation in high-throughput computational chemistry workflows. The workflow is validated through a structural screening of a data set of transition metal complexes extracted from the Cambridge Structural Database. To further illustrate the power and capabilities of MACE, we present the results of a model DFT study on the hemilability of pincer ligands in Ru, Fe, and Mn complexes, which highlights the utility of the workflow for both focused mechanistic studies and larger-scale high-throughput pipelines.

Classification of Hemilabile Ligands Using Machine Learning

Hemilabile ligands have the capacity to partially disengage from a metal center, providing a strategy to balance stability and reactivity in catalysis, but they are not straightforward to identify. We identify ligands in the Cambridge Structural Database that have been crystallized with distinct denticities and are thus identifiable as hemilabile ligands. We implement a semi-supervised learning approach using a label-spreading algorithm to augment a small negative set that is supported by heuristic rules of ligand and metal co-occurrence. We show that a heuristic based on coordinating atom identity alone is not sufficient to identify whether a ligand is hemilabile, and our trained machine-learning classification models are instead needed to predict whether a bi-, tri-, or tetradentate ligand is hemilabile with high accuracy and precision. Feature importance analysis of our models shows that the second, third, and fourth coordination spheres all play important roles in ligand hemilability.

HiREX: High-Throughput Reactivity Exploration for Extended Databases of Transition-Metal Catalysts

A method is introduced for the automated analysis of reactivity exploration for extended in silico databases of transition-metal catalysts. The proposed workflow is designed to tackle two key challenges for bias-free mechanistic explorations on large databases of catalysts: (1) automated exploration of the chemical space around each catalyst with unique structural and chemical features and (2) automated analysis of the resulting large chemical data sets. To address these challenges, we have extended the application of our previously developed ReNeGate method for bias-free reactivity exploration and implemented an automated analysis procedure to identify the classes of reactivity patterns within specific catalyst groups. Our procedure applied to an extended series of representative Mn(I) pincer complexes revealed correlations between structural and reactive features, pointing to new channels for catalyst transformation under the reaction conditions. Such an automated high-throughput virtual screening of systematically generated hypothetical catalyst data sets opens new opportunities for the design of high-performance catalysts as well as an accelerated method for expert bias-free high-throughput in silico reactivity exploration.

DFT-based analysis of siderophore-metal ion interaction for efficient heavy metal remediation

Siderophores have great application potential in metal pollutant remediation because of their effective cost and friendly impact on the environment. However, the practical use of siderophores in the remediation of specific metals is rather limited because of the weak nonspecific interactions between the siderophores and different metals. Thus, screening for a siderophore with optimal interaction with a specific metal would be necessary. In this study, the interaction between metal ions and moieties that donate the oxygen ligands for the coordination of four types of siderophore (hydroxamates, catecholates, phenolates, and carboxylates) was modeled and analyzed. As revealed by DFT-based analysis, the four types of siderophore generally exhibited selection preference for different metal ions in the order Ga3+  > Al3+  > Fe3+  > Cr3+  > Ni2+  > Cu2+  > Zn2+  > Co2+  > Mn2+  > Hg2+  > Pb2+  > Cd2+, which was determined mainly by the electronegativity of the siderophore functional groups, the electronegativity of the metals, and the ionic radius of the metals, as well as the interaction between the siderophores and the metals. Moreover, the effect of linear or nonlinear (cyclic) structure on the affinity of each siderophore for different metal ions was evaluated. In most situations, metal-bound cyclic siderophores were found to be more stable than their linear counterparts. Thus, proper siderophores for the remediation of metal pollution may be rapidly screened using this model.

ReNeGate: A Reaction Network Graph-Theoretical Tool for Automated Mechanistic Studies in Computational Homogeneous Catalysis

Exploration of the chemical reaction space of chemical transformations in multicomponent mixtures is one of the main challenges in contemporary computational chemistry. To remove expert bias from mechanistic studies and to discover new chemistries, an automated graph-theoretical methodology is proposed, which puts forward a network formalism of homogeneous catalysis reactions and utilizes a network analysis tool for mechanistic studies. The method can be used for analyzing trajectories with single and multiple catalytic species and can provide unique conformers of catalysts including multinuclear catalyst clusters along with other catalytic mixture components. The presented three-step approach has the integrated ability to handle multicomponent catalytic systems of arbitrary complexity (mixtures of reactants, catalyst precursors, ligands, additives, and solvents). It is not limited to predefined chemical rules, does not require prealignment of reaction mixture components consistent with a reaction coordinate, and is not agnostic to the chemical nature of transformations. Conformer exploration, reactive event identification, and reaction network analysis are the main steps taken for identifying the pathways in catalytic systems given the starting precatalytic reaction mixture as the input. Such a methodology allows us to efficiently explore catalytic systems in realistic conditions for either previously observed or completely unknown reactive events in the context of a network representing different intermediates. Our workflow for the catalytic reaction space exploration exclusively focuses on the identification of thermodynamically feasible conversion channels, representative of the (secondary) catalyst deactivation or inhibition paths, which are usually most difficult to anticipate based solely on expert chemical knowledge. Thus, the expert bias is sought to be removed at all steps, and the chemical intuition is limited to the choice of the thermodynamic constraint imposed by the applicable experimental conditions in terms of threshold energy values for allowed transformations. The capabilities of the proposed methodology have been tested by exploring the reactivity of Mn complexes relevant for catalytic hydrogenation chemistry to verify previously postulated activation mechanisms and unravel unexpected reaction channels relevant to rare deactivation events.

Prediction of protein pK a with representation learning

The behavior of proteins is closely related to the protonation states of the residues. Therefore, prediction and measurement of pK a are essential to understand the basic functions of proteins. In this work, we develop a new empirical scheme for protein pK a prediction that is based on deep representation learning. It combines machine learning with atomic environment vector (AEV) and learned quantum mechanical representation from ANI-2x neural network potential (J. Chem. Theory Comput. 2020, 16, 4192). The scheme requires only the coordinate information of a protein as the input and separately estimates the pK a for all five titratable amino acid types. The accuracy of the approach was analyzed with both cross-validation and an external test set of proteins. Obtained results were compared with the widely used empirical approach PROPKA. The new empirical model provides accuracy with MAEs below 0.5 for all amino acid types. It surpasses the accuracy of PROPKA and performs significantly better than the null model. Our model is also sensitive to the local conformational changes and molecular interactions.

ChemSpaX: exploration of chemical space by automated functionalization of molecular scaffold

Exploration of the local chemical space of molecular scaffolds by post-functionalization (PF) is a promising route to discover novel molecules with desired structure and function. PF with rationally chosen substituents based on known electronic and steric properties is a commonly used experimental and computational strategy in screening, design and optimization of catalytic scaffolds. Automated generation of reasonably accurate geometric representations of post-functionalized molecular scaffolds is highly desirable for data-driven applications. However, automated PF of transition metal (TM) complexes remains challenging. In this work a Python-based workflow, ChemSpaX, that is aimed at automating the PF of a given molecular scaffold with special emphasis on TM complexes, is introduced. In three representative applications of ChemSpaX by comparing with DFT and DFT-B calculations, we show that the generated structures have a reasonable quality for use in computational screening applications. Furthermore, we show that ChemSpaX generated geometries can be used in machine learning applications to accurately predict DFT computed HOMO-LUMO gaps for transition metal complexes. ChemSpaX is open-source and aims to bolster and democratize the efforts of the scientific community towards data-driven chemical discovery.

Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning

Transition-metal complexes are attractive targets for the design of catalysts and functional materials. The behavior of the metal-organic bond, while very tunable for achieving target properties, is challenging to predict and necessitates searching a wide and complex space to identify needles in haystacks for target applications. This review will focus on the techniques that make high-throughput search of transition-metal chemical space feasible for the discovery of complexes with desirable properties. The review will cover the development, promise, and limitations of "traditional" computational chemistry (i.e., force field, semiempirical, and density functional theory methods) as it pertains to data generation for inorganic molecular discovery. The review will also discuss the opportunities and limitations in leveraging experimental data sources. We will focus on how advances in statistical modeling, artificial intelligence, multiobjective optimization, and automation accelerate discovery of lead compounds and design rules. The overall objective of this review is to showcase how bringing together advances from diverse areas of computational chemistry and computer science have enabled the rapid uncovering of structure-property relationships in transition-metal chemistry. We aim to highlight how unique considerations in motifs of metal-organic bonding (e.g., variable spin and oxidation state, and bonding strength/nature) set them and their discovery apart from more commonly considered organic molecules. We will also highlight how uncertainty and relative data scarcity in transition-metal chemistry motivate specific developments in machine learning representations, model training, and in computational chemistry. Finally, we will conclude with an outlook of areas of opportunity for the accelerated discovery of transition-metal complexes.

Trends in computational molecular catalyst design

Computational methods have emerged as a powerful tool to augment traditional experimental molecular catalyst design by providing useful predictions of catalyst performance and decreasing the time needed for catalyst screening. In this perspective, we discuss three approaches for computational molecular catalyst design: (i) the reaction mechanism-based approach that calculates all relevant elementary steps, finds the rate and selectivity determining steps, and ultimately makes predictions on catalyst performance based on kinetic analysis, (ii) the descriptor-based approach where physical/chemical considerations are used to find molecular properties as predictors of catalyst performance, and (iii) the data-driven approach where statistical analysis as well as machine learning (ML) methods are used to obtain relationships between available data/features and catalyst performance. Following an introduction to these approaches, we cover their strengths and weaknesses and highlight some recent key applications. Furthermore, we present an outlook on how the currently applied approaches may evolve in the near future by addressing how recent developments in building automated computational workflows and implementing advanced ML models hold promise for reducing human workload, eliminating human bias, and speeding up computational catalyst design at the same time. Finally, we provide our viewpoint on how some of the challenges associated with the up-and-coming approaches driven by automation and ML may be resolved.