ASKCOS: an open source software suite for synthesis planning
🤖 AI Summary
The field of chemical synthesis route planning lacks efficient, open-source, multimodal computational tools. Method: This work introduces the first integrated, open-source software suite for retrosynthetic analysis, encompassing reaction condition recommendation, product prediction, solubility estimation, and quantum-mechanical descriptor computation (e.g., solvation free energy, pKa). It unifies state-of-the-art single-step retrosynthesis models (Transformer- and GNN-based), reaction outcome classifiers, condition prediction models, and machine learning–based physical property predictors within a unified framework, augmented by a multi-dimensional feasibility assessment module. Contribution/Results: The suite establishes a closed-loop decision-support framework—from retrosynthetic proposal to experimental executability validation—demonstrating significant improvements in route design efficiency and success rate. Deployed across academia and industry, it has been adopted by hundreds of medicinal chemists and university researchers, emerging as one of the most widely used computer-aided synthesis (CAS) platforms.
Application Category
📝 Abstract
The advancement of machine learning and the availability of large-scale reaction datasets have accelerated the development of data-driven models for computer-aided synthesis planning (CASP) in the past decade. Here, we detail the newest version of ASKCOS, an open source software suite for synthesis planning that makes available several research advances in a freely available, practical tool. Four one-step retrosynthesis models form the basis of both interactive planning and automatic planning modes. Retrosynthetic planning is complemented by other modules for feasibility assessment and pathway evaluation, including reaction condition recommendation, reaction outcome prediction, and auxiliary capabilities such as solubility prediction and quantum mechanical descriptor prediction. ASKCOS has assisted hundreds of medicinal, synthetic, and process chemists in their day-to-day tasks, complementing expert decision making. It is our belief that CASP tools like ASKCOS are an important part of modern chemistry research, and that they offer ever-increasing utility and accessibility.
Problem
Research questions and friction points this paper is trying to address.
Machine Learning
Chemical Synthesis Prediction
Dissolution Predictability
Innovation
Methods, ideas, or system contributions that make the work stand out.
Machine Learning
Chemical Synthesis Planning
User-Friendly Software
🔎 Similar Papers
No similar papers found.
