Locally Optimizable Joint Embedding Framework to Design Nitrogen‐rich Molecules that are Similar but Improved. Issue 7 (28th April 2021)

• Record Type: Journal Article
• Authors: Balakrishnan, Sangeeth; VanGessel, Francis G.; Boukouvalas, Zois; Barnes, Brian C.; Fuge, Mark D.; Chung, Peter W.
• Is Part Of: Molecular informatics. Volume 40:Issue 7(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-04-28
• Subjects: nitrogen-rich molecules -- molecule optimization -- machine learning -- generative modeling -- Chapman-Jouguet velocity; Cheminformatics -- Periodicals; QSAR (Biochemistry) -- Periodicals; Structure-activity relationships (Biochemistry) -- Periodicals; Drugs -- Structure-activity relationships -- Periodicals; 615.19
• DOI: 10.1002/minf.202100011
• Languages: English
• ISSNs: 1868-1743

Abstract

Abstract: Deep learning has shown great potential for generating molecules with desired properties. But the cost and time required to obtain relevant property data have limited study to only a few classes of materials for which extensive data have already been collected. We develop a deep learning method that combines a generative model with a property prediction model to fuse small data of one class of molecules with larger data in another class. Common low‐level physicochemical properties are jointly embedded into a latent space that can be used to design molecules in the smaller class. The chemical space around the molecules in the training set is explored through local gradient ascent optimization. Based on nine molecules from the original training set, nine new molecules are found to have improved properties while remaining structurally similar to the training molecules thereby easing requirements for entirely new synthesis routes. Validation is performed using an equilibrium thermochemistry code to verify the molecules and target properties. A specific example targeting the Chapman‐Jouguet velocity and small data for nitrogen‐rich molecules is shown. Despite the relative lack of nitrogen‐rich molecule data, the results demonstrate that fusing and joint embedding with plentiful low nitrogen molecular data can produce higher generative performance than using the scarce data alone. Abstract :