ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations

Ksenia Korovina, Sailun Xu, Kirthevasan Kandasamy, Willie Neiswanger, Barnabas Poczos, Jeff Schneider, Eric Xing
; Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, PMLR 108:3393-3403, 2020.

Abstract

In applications such as molecule design or drug discovery, it is desirable to have an algorithm which recommends new candidate molecules based on the results of past tests. These molecules first need to be synthesized and then tested for objective properties. We describe ChemBO, a Bayesian optimization framework for generating and optimizing organic molecules for desired molecular properties. While most existing data-driven methods for this problem do not account for sample efficiency or fail to enforce realistic constraints on synthesizability, our approach explores synthesis graphs in a sample-efficient way and produces synthesizable candidates. We implement ChemBO as a Gaussian process model and explore existing molecular kernels for it. Moreover, we propose a novel optimal-transport based distance and kernel that accounts for graphical information explicitly. In our experiments, we demonstrate the efficacy of the proposed approach on several molecular optimization problems.

Cite this Paper


BibTeX
@InProceedings{pmlr-v108-korovina20a, title = {ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations}, author = {Korovina, Ksenia and Xu, Sailun and Kandasamy, Kirthevasan and Neiswanger, Willie and Poczos, Barnabas and Schneider, Jeff and Xing, Eric}, pages = {3393--3403}, year = {2020}, editor = {Silvia Chiappa and Roberto Calandra}, volume = {108}, series = {Proceedings of Machine Learning Research}, address = {Online}, month = {26--28 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v108/korovina20a/korovina20a.pdf}, url = {http://proceedings.mlr.press/v108/korovina20a.html}, abstract = {In applications such as molecule design or drug discovery, it is desirable to have an algorithm which recommends new candidate molecules based on the results of past tests. These molecules first need to be synthesized and then tested for objective properties. We describe ChemBO, a Bayesian optimization framework for generating and optimizing organic molecules for desired molecular properties. While most existing data-driven methods for this problem do not account for sample efficiency or fail to enforce realistic constraints on synthesizability, our approach explores synthesis graphs in a sample-efficient way and produces synthesizable candidates. We implement ChemBO as a Gaussian process model and explore existing molecular kernels for it. Moreover, we propose a novel optimal-transport based distance and kernel that accounts for graphical information explicitly. In our experiments, we demonstrate the efficacy of the proposed approach on several molecular optimization problems.} }
Endnote
%0 Conference Paper %T ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations %A Ksenia Korovina %A Sailun Xu %A Kirthevasan Kandasamy %A Willie Neiswanger %A Barnabas Poczos %A Jeff Schneider %A Eric Xing %B Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2020 %E Silvia Chiappa %E Roberto Calandra %F pmlr-v108-korovina20a %I PMLR %J Proceedings of Machine Learning Research %P 3393--3403 %U http://proceedings.mlr.press %V 108 %W PMLR %X In applications such as molecule design or drug discovery, it is desirable to have an algorithm which recommends new candidate molecules based on the results of past tests. These molecules first need to be synthesized and then tested for objective properties. We describe ChemBO, a Bayesian optimization framework for generating and optimizing organic molecules for desired molecular properties. While most existing data-driven methods for this problem do not account for sample efficiency or fail to enforce realistic constraints on synthesizability, our approach explores synthesis graphs in a sample-efficient way and produces synthesizable candidates. We implement ChemBO as a Gaussian process model and explore existing molecular kernels for it. Moreover, we propose a novel optimal-transport based distance and kernel that accounts for graphical information explicitly. In our experiments, we demonstrate the efficacy of the proposed approach on several molecular optimization problems.
APA
Korovina, K., Xu, S., Kandasamy, K., Neiswanger, W., Poczos, B., Schneider, J. & Xing, E.. (2020). ChemBO: Bayesian Optimization of Small Organic Molecules with Synthesizable Recommendations. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, in PMLR 108:3393-3403

Related Material