We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

physics.chem-ph

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Physics > Chemical Physics

Title: Retroformer: Pushing the Limits of Interpretable End-to-end Retrosynthesis Transformer

Abstract: Retrosynthesis prediction is one of the fundamental challenges in organic synthesis. The task is to predict the reactants given a core product. With the advancement of machine learning, computer-aided synthesis planning has gained increasing interest. Numerous methods were proposed to solve this problem with different levels of dependency on additional chemical knowledge. In this paper, we propose Retroformer, a novel Transformer-based architecture for retrosynthesis prediction without relying on any cheminformatics tools for molecule editing. Via the proposed local attention head, the model can jointly encode the molecular sequence and graph, and efficiently exchange information between the local reactive region and the global reaction context. Retroformer reaches the new state-of-the-art accuracy for the end-to-end template-free retrosynthesis, and improves over many strong baselines on better molecule and reaction validity. In addition, its generative procedure is highly interpretable and controllable. Overall, Retroformer pushes the limits of the reaction reasoning ability of deep generative models.
Subjects: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG)
Cite as: arXiv:2201.12475 [physics.chem-ph]
  (or arXiv:2201.12475v1 [physics.chem-ph] for this version)

Submission history

From: Yue Wan [view email]
[v1] Sat, 29 Jan 2022 02:03:55 GMT (3396kb,D)

Link back to: arXiv, form interface, contact.