DirectMultiStep: Direct Route Generation for Multistep Retrosynthesis

Abstract

Traditional computer-aided synthesis planning (CASP) methods rely on iterative single-step predictions, leading to exponential search space growth that limits efficiency and scalability. We introduce a series of transformer-based models, that leverage a mixture of experts approach to directly generate multistep synthetic routes as a single string, conditionally predicting each transformation based on all preceding ones. Our DMS Explorer XL model, which requires only target compounds as input, outperforms state-of-the-art methods on the PaRoutes dataset with 1.9x and 3.1x improvements in Top-1 accuracy on the n${}_1$ and n${}_5$ test sets, respectively. Providing additional information, such as the desired number of steps and starting materials, enables both a reduction in model size and an increase in accuracy, highlighting the benefits of incorporating more constraints into the prediction process. The top-performing DMS-Flex (Duo) model scores 25-50% higher on Top-1 and Top-10 accuracies for both n${}_1$ and n${}_5$ sets. Additionally, our models successfully predict routes for FDA-approved drugs not included in the training data, demonstrating strong generalization capabilities. While the limited diversity of the training set may affect performance on less common reaction types, our multistep-first approach presents a promising direction towards fully automated retrosynthetic planning.

References in corpus (14)

• Decoupled Weight Decay Regularization
• Gaussian Error Linear Units (GELUs)
• Learning to Plan Chemical Syntheses
• State-of-the-Art Augmented NLP Transformer models for direct and single-step retrosynthesis
• Mixtral of Experts
• MechRetro is a chemical-mechanism-driven graph learning framework for interpretable retrosynthesis prediction and pathway planning
• Root-aligned SMILES: A Tight Representation for Chemical Reaction Prediction
• A Graph to Graphs Framework for Retrosynthesis Prediction
• Learning Graph Models for Retrosynthesis Prediction
• Learning to Make Generalizable and Diverse Predictions for Retrosynthesis
• A Bayesian algorithm for retrosynthesis
• RetroGraph: Retrosynthetic Planning with Graph Search
• RetCL: A Selection-based Approach for Retrosynthesis via Contrastive Learning
• RetroXpert: Decompose Retrosynthesis Prediction like a Chemist

Cited by in corpus (2)

• Quantum Machine Learning in Drug Discovery: Applications in Academia and Pharmaceutical Industries
• FragmentRetro: A Quadratic Retrosynthetic Method Based on Fragmentation Algorithms