A deep learning framework for accurate reaction prediction and its application on high-throughput experimentation data.

In recent years, it has been seen that artificial intelligence (AI) starts to bring revolutionary changes to chemical synthesis. However, the lack of suitable ways of representing chemical reactions and the scarceness of reaction data has limited the wider application of AI to reaction prediction. Here, we introduce a novel reaction representation, GraphRXN, for reaction prediction. It utilizes a universal graph-based neural network framework to encode chemical reactions by directly taking two-dimension reaction structures as inputs. The GraphRXN model was evaluated by three publically available chemical reaction datasets and gave on-par or superior results compared with other baseline models. To further evaluate the effectiveness of GraphRXN, wet-lab experiments were carried out for the purpose of generating reaction data. GraphRXN model was then built on high-throughput experimentation data and a decent accuracy (R2 of 0.712) was obtained on our in-house data. This highlights that the GraphRXN model can be deployed in an integrated workflow which combines robotics and AI technologies for forward reaction prediction.

HTE- and AI-assisted development of DHP-catalyzed decarboxylative selenation.

1,4-Dihydropyridine (DHP) derivatives play key roles in biology, but are rarely used as catalysts in synthesis. Here, we developed a DHP derivative-catalyzed decarboxylative selenation reaction that showed a broad substrate scope, with the assistance of high-throughput experimentation (HTE) and artificial intelligence (AI). The AI-based model could identify the key structural features and give accurate prediction of unseen reactions (R2 = 0.89, RMSE = 9.0%, and MAE = 6.3%). Our work not only developed the catalytic applications of DHP derivatives, but also demonstrated the power of the combination of HTE and AI to advance chemical synthesis.

Predicting the Feasibility of Copper(I)-Catalyzed Alkyne-Azide Cycloaddition Reactions Using a Recurrent Neural Network with a Self-Attention Mechanism.

The copper(I)-catalyzed alkyne-azide cycloaddition (CuAAC) reaction, a major click chemistry reaction, is widely employed in drug discovery and chemical biology. However, the success rate of the CuAAC reaction is not satisfactory as expected, and in order to improve its performance, we developed a recurrent neural network (RNN) model to predict its feasibility. First, we designed and synthesized a structurally diverse library of 700 compounds with the CuAAC reaction to obtain experimental data. Then, using reaction SMILES as input, we generated a bidirectional long-short-term memory with a self-attention mechanism (BiLSTM-SA) model. Our best prediction model has total accuracy of 80%. With the self-attention mechanism, adverse substructures responsible for negative reactions were recognized and derived as quantitative descriptors. Density functional theory investigations were conducted to provide evidence for the correlation between bromo-α-C hybrid types and the success rate of the reaction. Quantitative descriptors combined with RDKit descriptors were fed to three machine learning models, a support vector machine, random forest, and logistic regression, and resulted in improved performance. The BiLSTM-SA model for predicting the feasibility of the CuAAC reaction is superior to other conventional learning methods and advances heuristic chemical rules.

SyntaLinker: automatic fragment linking with deep conditional transformer neural networks.

Linking fragments to generate a focused compound library for a specific drug target is one of the challenges in fragment-based drug design (FBDD). Hereby, we propose a new program named SyntaLinker, which is based on a syntactic pattern recognition approach using deep conditional transformer neural networks. This state-of-the-art transformer can link molecular fragments automatically by learning from the knowledge of structures in medicinal chemistry databases (e.g. ChEMBL database). Conventionally, linking molecular fragments was viewed as connecting substructures that were predefined by empirical rules. In SyntaLinker, however, the rules of linking fragments can be learned implicitly from known chemical structures by recognizing syntactic patterns embedded in SMILES notations. With deep conditional transformer neural networks, SyntaLinker can generate molecular structures based on a given pair of fragments and additional restrictions. Case studies have demonstrated the advantages and usefulness of SyntaLinker in FBDD.

Identifying Novel Anti-Osteoporosis Leads with a Chemotype-Assembly Approach.

In this paper, we applied a chemotype-assembly approach for ligand-based drug discovery (LBDD) to discover novel anti-osteoporosis leads. With this new approach, we identified 12 chemotypes and derived 18 major chemotype assembly rules from 245 known anti-osteoporosis compounds. Then, we selected 19 compounds from an in-house compound library using chemotype-assembly approach for anti-osteoporosis assays, which resulted in 13 hits. Based on structural features in these 13 compounds, we synthesized 50 possible anti-osteoporosis compounds from the anti-osteoporosis chemotypes by means of click chemistry techniques and discovered a compound (10a, IC50 = 2 nM) with nanomolar activity. Compound 10a was then proved to be an anti-osteoporosis lead since it can prevent bone loss in vivo.