Machine Learning to Develop Peptide Catalysts-Successes, Limitations, and Opportunities.

Peptides have been established as modular catalysts for various transformations. Still, the vast number of potential amino acid building blocks renders the identification of peptides with desired catalytic activity challenging. Here, we develop a machine-learning workflow for the optimization of peptide catalysts. First-in a hypothetical competition-we challenged our workflow to identify peptide catalysts for the conjugate addition reaction of aldehydes to nitroolefins and compared the performance of the predicted structures with those optimized in our laboratory. On the basis of the positive results, we established a universal training set (UTS) containing 161 catalysts to sample an in silico library of ∼30,000 tripeptide members. Finally, we challenged our machine learning strategy to identify a member of the library as a stereoselective catalyst for an annulation reaction that has not been catalyzed by a peptide thus far. We conclude with a comparison of data-driven versus expert-knowledge-guided peptide catalyst optimization.

Effects of Ring Size and Steric Encumbrance on Boron-to-Palladium Transmetalation from Arylboronic Esters.

The structure of the diol from which an arylboronic ester is derived dramatically influences the rate of transmetalation in the Suzuki-Miyaura cross-coupling reaction. Some esters undergo transmetalation more than 20 times faster than the parent arylboronic acid. Herein, investigations into the influence of arylboronic ester ring size and steric properties on the mechanism of transmetalation in the Suzuki-Miyaura reaction are described. Both factors impact the propensity of an arylboronic ester to bind to a dimeric palladium hydroxide complex. The reaction of hindered arylboronic esters derived from 1,2-diols (1,3,2-dioxaborolanes) with palladium hydroxide dimers to form a complex incorporating a Pd-O-B linkage is thermodynamically favorable, but the barrier to coordination is often higher than the barrier to arene transfer. In contrast, the analogous reaction between arylboronic esters derived from 1,3-diols (1,3,2-dioxaborinanes) and palladium hydroxide dimers is thermodynamically unfavorable, as 1,3,2-dioxaborinanes exhibit decreased electrophilicity compared to esters derived from 1,2- or 1,4-diols. These factors also influence the barrier of the arene transfer step, and in many cases, arylboronic esters that do not easily form Pd-O-B linked complexes undergo transmetalation faster than those that do because of hyperconjugative stabilization of the arene transfer transition state.

A machine-learning tool to predict substrate-adaptive conditions for Pd-catalyzed C-N couplings.

Machine-learning methods have great potential to accelerate the identification of reaction conditions for chemical transformations. A tool that gives substrate-adaptive conditions for palladium (Pd)-catalyzed carbon-nitrogen (C-N) couplings is presented. The design and construction of this tool required the generation of an experimental dataset that explores a diverse network of reactant pairings across a set of reaction conditions. A large scope of C-N couplings was actively learned by neural network models by using a systematic process to design experiments. The models showed good performance in experimental validation: Ten products were isolated in more than 85% yield from a range of couplings with out-of-sample reactants designed to challenge the models. Importantly, the developed workflow continually improves the prediction capability of the tool as the corpus of data grows.

Machine-Learning-Guided Discovery of Electrochemical Reactions.

The molecular structures synthesizable by organic chemists dictate the molecular functions they can create. The invention and development of chemical reactions are thus critical for chemists to access new and desirable functional molecules in all disciplines of organic chemistry. This work seeks to expedite the exploration of emerging areas of organic chemistry by devising a machine-learning-guided workflow for reaction discovery. Specifically, this study uses machine learning to predict competent electrochemical reactions. To this end, we first develop a molecular representation that enables the production of general models with limited training data. Next, we employ automated experimentation to test a large number of electrochemical reactions. These reactions are categorized as competent or incompetent mixtures, and a classification model was trained to predict reaction competency. This model is used to screen 38,865 potential reactions in silico, and the predictions are used to identify a number of reactions of synthetic or mechanistic interest, 80% of which are found to be competent. Additionally, we provide the predictions for the 38,865-member set in the hope of accelerating the development of this field. We envision that adopting a workflow such as this could enable the rapid development of many fields of chemistry.

Leveraging Machine Learning for Enantioselective Catalysis: From Dream to Reality.

Catalyst optimization for enantioselective transformations has traditionally relied on empirical evaluation of catalyst properties. Although this approach has been successful in the past it is intrinsically limited and inefficient. To address this problem, our laboratory has developed a fully informatics guided workflow to leverage the power of artificial intelligence (AI) and machine learning (ML) to accelerate the discovery and optimization of any class of catalyst for any transformation. This approach is mechanistically agnostic, but also serves as a discovery platform to identify high performing catalysts that can be subsequently investigated with physical organic methods to identify the origins of selectivity.

Dreams, False Starts, Dead Ends, and Redemption: A Chronicle of the Evolution of a Chemoinformatic Workflow for the Optimization of Enantioselective Catalysts.

Catalyst design in enantioselective catalysis has historically been driven by empiricism. In this endeavor, experimentalists attempt to qualitatively identify trends in structure that lead to a desired catalyst function. In this body of work, we lay the groundwork for an improved, alternative workflow that uses quantitative methods to inform decision making at every step of the process. At the outset, we define a library of synthetically accessible permutations of a catalyst scaffold with the philosophy that the library contains every potential catalyst we are willing to make. To represent these chiral molecules, we have developed general 3D representations, which can be calculated for tens of thousands of structures. This defines the total chemical space of a given catalyst scaffold; it is constructed on the basis of catalyst structure only without regard to a specific reaction or mechanism. As such, any algorithmic subset selection method, which is unsupervised (i.e., only considers catalyst structure), should provide an ideal initial screening set for any new reaction that can be catalyzed by that scaffold. Notably, because this design strategy, the same set of catalysts can be used for any reaction that can be catalyzed with that parent catalyst scaffold. These are tested experimentally, and statistical learning tools can be used to create a model relating catalyst structure to catalyst function. Further, this model can be used to predict the performance of each catalyst candidate in the greater database of virtual catalyst candidates. In this way, it is possible estimate the performance of tens of thousands of catalysts by experimentally testing a smaller subset. Using error assessment metrics, it is possible to understand the confidence in new predictions. An experimentalist using this tool can balance the predicted results (reward) with the prediction confidence (risk) when deciding which catalysts to synthesize next in an optimization campaign. These catalysts are synthesized and tested experimentally. At this stage, either the optimization is a success or the predicted values were incorrect and further optimization is required. In the case of the latter, the information can be fed back into the statistical learning model to refine the model, and this iterative process can be used to determine the optimal catalyst. In this body of work, we not only establish this workflow but quantitatively establish how best to execute each step. Herein, we evaluate several 3D molecular representations to determine how best to represent molecules. Several selection protocols are examined to best decide which set of molecules can be used to represent the library of interest. In addition, the number of reactions needed to make accurate, statistical learning models is evaluated. Taken together these components establish a tool ready to progress from the development stage to the utility stage. As such, current research endeavors focus on applying these tools to optimize new reactions.

Cautionary Guidelines for Machine Learning Studies with Combinatorial Datasets.

Regression modeling is becoming increasingly prevalent in organic chemistry as a tool for reaction outcome prediction and mechanistic interrogation. Frequently, to acquire the requisite amount of data for such studies, researchers employ combinatorial datasets to maximize the number of data points while limiting the number of discrete chemical entities required. An often-overlooked problem in modeling studies using combinatorial datasets is the tendency to fit on patterns in the datasets (i.e., the presence or absence of a reactant or catalyst) rather than to identify meaningful trends between descriptors and the response variable. Consequently, the generality and interpretability of such models suffer. This report illustrates these well-known pitfalls in a case study, demonstrates the necessary control experiments to identify when this property will be problematic, and suggests how to perform further validation to assess general applicability and interpretability of models trained using combinatorial datasets.

Development of a Computer-Guided Workflow for Catalyst Optimization. Descriptor Validation, Subset Selection, and Training Set Analysis.

Modern, enantioselective catalyst development is driven largely by empiricism. Although this approach has fostered the introduction of most of the existing synthetic methods, it is inherently limited by the skill, creativity, and chemical intuition of the practitioner. Herein, we present a complementary approach to catalyst optimization in which statistical methods are used at each stage to streamline development. To construct the optimization informatics workflow, a number of critical components had to be subjected to rigorous validation. First, the critically important molecular descriptors were validated in two case studies to establish the importance of conformation-dependent molecular representations. Next, with a large data set available, it was possible to investigate the amount of data necessary to make predictive models with different modeling methods. Given the commercial availability of many catalyst structures, it was possible to compare models generated with algorithmically selected training sets and commercially available training sets. Finally, the augmentation of limited data sets is demonstrated in a method informed by unsupervised learning to restore the accuracy of the generated models.

Quantitative Structure-Selectivity Relationships in Enantioselective Catalysis: Past, Present, and Future.

The dawn of the 21st century has brought with it a surge of research related to computer-guided approaches to catalyst design. In the past two decades, chemoinformatics, the application of informatics to solve problems in chemistry, has increasingly influenced prediction of activity and mechanistic investigations of organic reactions. The advent of advanced statistical and machine learning methods, as well as dramatic increases in computational speed and memory, has contributed to this emerging field of study. This review summarizes strategies to employ quantitative structure-selectivity relationships (QSSR) in asymmetric catalytic reactions. The coverage is structured by initially introducing the basic features of these methods. Subsequent topics are discussed according to increasing complexity of molecular representations. As the most applied subfield of QSSR in enantioselective catalysis, the application of local parametrization approaches and linear free energy relationships (LFERs) along with multivariate modeling techniques is described first. This section is followed by a description of global parametrization methods, the first of which is continuous chirality measures (CCM) because it is a single parameter derived from the global structure of a molecule. Chirality codes, global, multivariate descriptors, are then introduced followed by molecular interaction fields (MIFs), a global descriptor class that typically has the highest dimensionality. To highlight the current reach of QSSR in enantioselective transformations, a comprehensive collection of examples is presented. When combined with traditional experimental approaches, chemoinformatics holds great promise to predict new catalyst structures, rationalize mechanistic behavior, and profoundly change the way chemists discover and optimize reactions.

Prediction of higher-selectivity catalysts by computer-driven workflow and machine learning.

Catalyst design in asymmetric reaction development has traditionally been driven by empiricism, wherein experimentalists attempt to qualitatively recognize structural patterns to improve selectivity. Machine learning algorithms and chemoinformatics can potentially accelerate this process by recognizing otherwise inscrutable patterns in large datasets. Herein we report a computationally guided workflow for chiral catalyst selection using chemoinformatics at every stage of development. Robust molecular descriptors that are agnostic to the catalyst scaffold allow for selection of a universal training set on the basis of steric and electronic properties. This set can be used to train machine learning methods to make highly accurate predictive models over a broad range of selectivity space. Using support vector machines and deep feed-forward neural networks, we demonstrate accurate predictive modeling in the chiral phosphoric acid-catalyzed thiol addition to N-acylimines.

Evaluating continuous chirality measure as a 3D descriptor in chemoinformatics applied to asymmetric catalysis.

Continuous Chirality Measure (CCM) is a computational metric by which to quantify the chirality of a compound. In enantioselective catalysis, prior work has postulated that CCM is correlated to selectivity and can be used to understand which structural features dictate catalyst efficacy. Herein, the investigation of CCM as a metric capable of guiding catalyst optimization is explored. Conformer-dependent CCM is also explored. Finally, CCM is used with Sterimol parameters to significantly improve the performance of Random Forest models.

Elucidating the Role of the Boronic Esters in the Suzuki-Miyaura Reaction: Structural, Kinetic, and Computational Investigations.

The Suzuki-Miyaura reaction is the most practiced palladium-catalyzed, cross-coupling reaction because of its broad applicability, low toxicity of the metal (B), and the wide variety of commercially available boron substrates. A wide variety of boronic acids and esters, each with different properties, have been developed for this process. Despite the popularity of the Suzuki-Miyaura reaction, the precise manner in which the organic fragment is transferred from boron to palladium has remained elusive for these reagents. Herein, we report the observation and characterization of pretransmetalation intermediates generated from a variety of commonly employed boronic esters. The ability to confirm the intermediacy of pretransmetalation intermediates provided the opportunity to clarify mechanistic aspects of the transfer of the organic moiety from boron to palladium in the key transmetalation step. A series of structural, kinetic, and computational investigations revealed that boronic esters can transmetalate directly without prior hydrolysis. Furthermore, depending on the boronic ester employed, significant rate enhancements for the transfer of the B-aryl groups were observed. Overall, two critical features were identified that enable the transfer of the organic fragment from boron to palladium: (1) the ability to create an empty coordination site on the palladium atom and (2) the nucleophilic character of the ipso carbon bound to boron. Both of these features ultimately relate to the electron density of the oxygen atoms in the boronic ester.

Structural, Kinetic, and Computational Characterization of the Elusive Arylpalladium(II)boronate Complexes in the Suzuki-Miyaura Reaction.

The existence of the oft-invoked intermediates containing the crucial Pd-O-B subunit, the "missing link", has been established in the Suzuki-Miyaura cross-coupling reaction. The use of low-temperature, rapid injection NMR spectroscopy (RI-NMR), kinetic studies, and computational analysis has enabled the generation, observation, and characterization of these highly elusive species. The ability to confirm the intermediacy of Pd-O-B-containing species provided the opportunity to clarify mechanistic aspects of the transfer of the organic moiety from boron to palladium in the key transmetalation step. Specifically, these studies establish the identity of two different intermediates containing Pd-O-B linkages, a tri-coordinate (6-B-3) boronic acid complex and a tetra-coordinate (8-B-4) boronate complex, both of which undergo transmetalation leading to the cross-coupling product. Two distinct mechanistic pathways have been elucidated for stoichiometric reactions of these complexes: (1) transmetalation via an unactivated 6-B-3 intermediate that dominates in the presence of an excess of ligand, and (2) transmetalation via an activated 8-B-4 intermediate that takes place with a deficiency of ligand.