Bayesian molecular design with a chemical language model

Abstract

The aim of computational molecular design is the identification of promising hypothetical molecules with a predefined set of desired properties. We address the issue of accelerating the material discovery with state-of-the-art machine learning techniques. The method involves two different types of prediction; the forward and backward predictions. The objective of the forward prediction is to create a set of machine learning models on various properties of a given molecule. Inverting the trained forward models through Bayes’ law, we derive a posterior distribution for the backward prediction, which is conditioned by a desired property requirement. Exploring high-probability regions of the posterior with a sequential Monte Carlo technique, molecules that exhibit the desired properties can computationally be created. One major difficulty in the computational creation of molecules is the exclusion of the occurrence of chemically unfavorable structures. To circumvent this issue, we derive a chemical language model that acquires commonly occurring patterns of chemical fragments through natural language processing of ASCII strings of existing compounds, which follow the SMILES chemical language notation. In the backward prediction, the trained language model is used to refine chemical strings such that the properties of the resulting structures fall within the desired property region while chemically unfavorable structures are successfully removed. The present method is demonstrated through the design of small organic molecules with the property requirements on HOMO-LUMO gap and internal energy. The R package iqspr is available at the CRAN repository.Electronic supplementary materialThe online version of this article (doi:10.1007/s10822-016-0008-z) contains supplementary material, which is available to authorized users.

iQSPR in XenonPy: A Bayesian Molecular Design Algorithm

“…In iQSPR-X, a sequential Monte Carlo algorithm proposed by Ikebata et al [23] is implemented. Thus, P(S) will deliver a small or even zero probability when presented with an unfavorable or chemically unrealistic structure, thereby acting as a filter for such out-of-scope or invalid structures.…”

“…For the generator, the extended n-gram model developed by Ikebata et al [23] can be used by training it with any chemical structures given in SMILES. For the generator, the extended n-gram model developed by Ikebata et al [23] can be used by training it with any chemical structures given in SMILES.…”

“…Promising examples have included various types of varia-tional autoencoders, [12][13][14][15] generative adversarial networks, [16] recurrent neural networks, [17,18] and so on. Ikebata et al [23] combined a simple probabilistic language model based on an n-gram representation of SMILES sequences with a Bayesian inference framework to sequentially modify a population of molecules into promising candidate molecules that would exhibit desired properties. Datasets this large are unavailable in many applications.…”

“…[23] The algorithm was implemented in XenonPy, a Python package with an integrated platform of materials informatics. [23] The algorithm was implemented in XenonPy, a Python package with an integrated platform of materials informatics.…”

Hunting for Organic Molecules with Artificial Intelligence: Molecules Optimized for Desired Excitation Energies

“…It has been brought closer to reality by recent advances on machine learning algorithms for de novo molecule design, that do not need handcrafted chemical rules [1][2][3][4][5] . Figure 1 illustrates our AI-assisted chemistry platform to develop new molecules.…”