SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties

TL;DR

SMILES2vec is an interpretable deep neural network that predicts various chemical properties directly from SMILES strings, outperforming traditional methods and providing insights aligned with chemical principles.

Contribution

This work introduces SMILES2vec, a general-purpose, interpretable deep RNN model that predicts multiple chemical properties directly from SMILES without manual feature engineering.

Findings

Outperforms traditional MLP models with engineered features

Achieves 88% accuracy in identifying chemically relevant parts for solubility

Demonstrates generalization across toxicity, activity, solubility, and solvation energy

Abstract

Chemical databases store information in text representations, and the SMILES format is a universal standard used in many cheminformatics software. Encoded in each SMILES string is structural information that can be used to predict complex chemical properties. In this work, we develop SMILES2vec, a deep RNN that automatically learns features from SMILES to predict chemical properties, without the need for additional explicit feature engineering. Using Bayesian optimization methods to tune the network architecture, we show that an optimized SMILES2vec model can serve as a general-purpose neural network for predicting distinct chemical properties including toxicity, activity, solubility and solvation energy, while also outperforming contemporary MLP neural networks that uses engineered features. Furthermore, we demonstrate proof-of-concept of interpretability by developing an explanation mask that localizes on the most important characters used in making a prediction. When tested on the solubility dataset, it identified specific parts of a chemical that is consistent with established first-principles knowledge with an accuracy of 88%. Our work demonstrates that neural networks can learn technically accurate chemical concept and provide state-of-the-art accuracy, making interpretable deep neural networks a useful tool of relevance to the chemical industry.