Data-driven Chemical Reaction Prediction and Retrosynthesis.

The synthesis of organic compounds, which is central to many areas such as drug discovery, material synthesis and biomolecular chemistry, requires chemists to have years of knowledge and experience. The development of technologies with the potential to learn and support experts in the design of synthetic routes is a half-century-old challenge with an interesting revival in the last decade. In fact, the renewed interest in artificial intelligence (AI), driven mainly by data availability, is profoundly changing the landscape of computer-aided chemical reaction prediction and retrosynthetic analysis. In this article, we briefly review different approaches to predict forward reactions and retrosynthesis, with a strong focus on data-driven ones. While data-driven technologies still need to demonstrate their full potential compared to expert rule-based systems in synthetic chemistry, the acceleration experienced in the last decade is a convincing sign that where we use software today, there will be AI tomorrow. This revolution will help and empower bench chemists, driving the transformation of chemistry towards a high-tech business over the next decades.

Inferring experimental procedures from text-based representations of chemical reactions.

The experimental execution of chemical reactions is a context-dependent and time-consuming process, often solved using the experience collected over multiple decades of laboratory work or searching similar, already executed, experimental protocols. Although data-driven schemes, such as retrosynthetic models, are becoming established technologies in synthetic organic chemistry, the conversion of proposed synthetic routes to experimental procedures remains a burden on the shoulder of domain experts. In this work, we present data-driven models for predicting the entire sequence of synthesis steps starting from a textual representation of a chemical equation, for application in batch organic chemistry. We generated a data set of 693,517 chemical equations and associated action sequences by extracting and processing experimental procedure text from patents, using state-of-the-art natural language models. We used the attained data set to train three different models: a nearest-neighbor model based on recently-introduced reaction fingerprints, and two deep-learning sequence-to-sequence models based on the Transformer and BART architectures. An analysis by a trained chemist revealed that the predicted action sequences are adequate for execution without human intervention in more than 50% of the cases.

Automated extraction of chemical synthesis actions from experimental procedures.

Experimental procedures for chemical synthesis are commonly reported in prose in patents or in the scientific literature. The extraction of the details necessary to reproduce and validate a synthesis in a chemical laboratory is often a tedious task requiring extensive human intervention. We present a method to convert unstructured experimental procedures written in English to structured synthetic steps (action sequences) reflecting all the operations needed to successfully conduct the corresponding chemical reactions. To achieve this, we design a set of synthesis actions with predefined properties and a deep-learning sequence to sequence model based on the transformer architecture to convert experimental procedures to action sequences. The model is pretrained on vast amounts of data generated automatically with a custom rule-based natural language processing approach and refined on manually annotated samples. Predictions on our test set result in a perfect (100%) match of the action sequence for 60.8% of sentences, a 90% match for 71.3% of sentences, and a 75% match for 82.4% of sentences.

Predicting retrosynthetic pathways using transformer-based models and a hyper-graph exploration strategy.

We present an extension of our Molecular Transformer model combined with a hyper-graph exploration strategy for automatic retrosynthesis route planning without human intervention. The single-step retrosynthetic model sets a new state of the art for predicting reactants as well as reagents, solvents and catalysts for each retrosynthetic step. We introduce four metrics (coverage, class diversity, round-trip accuracy and Jensen-Shannon divergence) to evaluate the single-step retrosynthetic models, using the forward prediction and a reaction classification model always based on the transformer architecture. The hypergraph is constructed on the fly, and the nodes are filtered and further expanded based on a Bayesian-like probability. We critically assessed the end-to-end framework with several retrosynthesis examples from literature and academic exams. Overall, the frameworks have an excellent performance with few weaknesses related to the training data. The use of the introduced metrics opens up the possibility to optimize entire retrosynthetic frameworks by focusing on the performance of the single-step model only.