On AI-centered Retrosynthetic Planning: a Survey

TMLR Paper1949 Authors

14 Dec 2023 (modified: 17 Sept 2024)Rejected by TMLREveryoneRevisionsBibTeXCC BY 4.0
Abstract: Retrosynthetic planning is one of the most challenging problems in organic chemistry up to date. It involves designing a target molecule by recursively decomposing it into simpler molecules through a series of backward chemical reaction steps. Typically, these simpler molecules are either commercially available or easy to synthesize. The development of new computer-aided retrosynthetic methods can advance the development of target molecules in drug design, material science and agrochemicals. This paper presents the retrosynthetic planning problem and the current AI-based methods. We conduct an in-depth review of the application of AI-based methods to retrosynthetic planning while also presenting the current challenges and potential future research directions.
Submission Length: Regular submission (no more than 12 pages of main content)
Assigned Action Editor: ~Stanislaw_Kamil_Jastrzebski1
Submission Number: 1949
Loading