Learning Molecular Representations for Thermochemistry Prediction of Cyclic Hydrocarbons and Oxygenates.

Accurate thermochemistry estimation of polycyclic molecules is crucial for kinetic modeling of chemical processes that use renewable and alternative feedstocks. In kinetic model generators, molecular properties are estimated rapidly with group additivity, but this method is known to have limitations for polycyclic structures. This issue has been resolved in our work by combining a geometry-based molecular representation with a deep neural network trained on ab initio data. Each molecule is transformed into a probabilistic vector from its interatomic distances, bond angles, and dihedral angles. The model is tested on a small experimental dataset (200 molecules) from the literature, a new medium-sized set (4000 molecules) with both open-shell and closed-shell species, calculated at the CBS-QB3 level with empirical corrections, and a large G4MP2-level QM9-based dataset (40 000 molecules). Heat capacities between 298.15 and 2500 K are calculated in the medium set with an average deviation of about 1.5 J mol-1 K-1 and the standard entropy at 298.15 K is predicted with an average error below 4 J mol-1 K-1. The standard enthalpy of formation at 298.15 K has an average out-of-sample error below 4 kJ mol-1 on a QM9 training set size of around 15 000 molecules. By fitting NASA polynomials, the enthalpy of formation at higher temperatures can be calculated with the same accuracy as the standard enthalpy of formation. Uncertainty quantification by means of the ensemble standard deviation is included to indicate when molecules that are on the edge or outside of the application range of the model are evaluated.

A robotic platform for flow synthesis of organic compounds informed by AI planning.

The synthesis of complex organic molecules requires several stages, from ideation to execution, that require time and effort investment from expert chemists. Here, we report a step toward a paradigm of chemical synthesis that relieves chemists from routine tasks, combining artificial intelligence-driven synthesis planning and a robotically controlled experimental platform. Synthetic routes are proposed through generalization of millions of published chemical reactions and validated in silico to maximize their likelihood of success. Additional implementation details are determined by expert chemists and recorded in reusable recipe files, which are executed by a modular continuous-flow platform that is automatically reconfigured by a robotic arm to set up the required unit operations and carry out the reaction. This strategy for computer-augmented chemical synthesis is demonstrated for 15 drug or drug-like substances.

Automated reaction database and reaction network analysis: extraction of reaction templates using cheminformatics.

Both the automated generation of reaction networks and the automated prediction of synthetic trees require, in one way or another, the definition of possible transformations a molecule can undergo. One way of doing this is by using reaction templates. In view of the expanding amount of known reactions, it has become more and more difficult to envision all possible transformations that could occur in a studied system. Nonetheless, most reaction network generation tools rely on user-defined reaction templates. Not only does this limit the amount of chemistry that can be accounted for in the reaction networks, it also confines the wide-spread use of the tools by a broad public. In retrosynthetic analysis, the quality of the analysis depends on what percentage of the known chemistry is accounted for. Using databases to identify templates is therefore crucial in this respect. For this purpose, an algorithm has been developed to extract reaction templates from various types of chemical databases. Some databases such as the Kyoto Encyclopedia for Genes and Genomes and RMG do not report an atom-atom mapping (AAM) for the reactions. This makes the extraction of a template non-straightforward. If no mapping is available, it is calculated by the Reaction Decoder Tool (RDT). With a correct AAM-either calculated by RDT or specified-the algorithm consistently extracts a correct template for a wide variety of reactions, both elementary and non-elementary. The developed algorithm is a first step towards data-driven generation of synthetic trees or reaction networks, and a greater accessibility for non-expert users.