In this poster, I will summarize our development of the open access ASKCOS synthesis planning suite (askcos.mit.edu). Advances in laboratory automation promise to decrease the manual effort of synthesis, but determining how to synthesize a compound currently requires time and effort investment from expert chemists. Our overall synthesis planning workflow contains a number of interconnected modules. First, we address the problem of retrosynthetic planning and how the recursive expansion and search strategy are both conducive to machine learning approaches. Second, we address the challenge of in silico reaction validation, which can be addressed by solving the inverse problem of forward reaction prediction. This poster will describe how our techniques for retrosynthesis and forward prediction—leveraging advances in data science and machine learning techniques—are integrated into an overall workflow that, for a given molecular target, predicts a rank-ordered list of reaction paths that connect the target to purchasable starting materials via a series of plausible reaction steps. The software is able to find robust synthetic pathways in seconds for simple organic molecules and in less than a minute even for modern small molecule APIs. I will also discuss how our software has been adopted by a medicinal and process chemistry groups in several pharmaceutical and chemical companies and opportunities for extension.

ASKCOS: Data-driven synthetic route design and validation for small organic molecules

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.