Deep Architectures and Deep Learning in Chemoinformatics: The Prediction of Aqueous Solubility for Drug-Like Molecules

Abstract

Shallow machine learning methods have been applied to chemoinformatics problems with some success. As more data becomes available and more complex problems are tackled, deep machine learning methods may also become useful. Here we present a brief overview of deep learning methods and show in particular how recursive neural network approaches can be applied to the problem of predicting molecular properties. However molecules are typically described by undirected cyclic graphs, while recursive approaches typically use directed acyclic graphs. Thus we develop methods to address this discrepancy, essentially by considering an ensemble of recursive neural networks associated with all possible vertex-centered acyclic orientations of the molecular graph. One advantage of this approach is that it relies only minimally on the identification of suitable molecular descriptors, since suitable representations are learnt automatically from the data. Several variants of this approach are applied to the problem of predicting aqueous solubility and tested on four benchmark datasets. Experimental results show that the performance of the deep learning methods matches or exceeds the performance of other state-of-the-art methods according to several evaluation metrics and expose the fundamental limitations arising from training sets that are too small or too noisy. A web-based predictor AquaSol is available online through the ChemDB portal (cdb.ics.uci.edu) together with additional material.

Deep Learning for Drug-Induced Liver Injury

“…It has been confirmed that deep-learning architectures have the power to handle big data with little manual intervention. 37,38 These practical and useful techniques also have been applied to chemoinformatics and bioinformatics [39][40][41][42][43][44][45][46][47][48][49] for dealing with different tasks, such as, aqueous solubility prediction, quantitative structure-activity relationship analysis, 48 and predicting the sequence specificities of DNA/RNA binding proteins. 50 Motivated by its great success in these fields mentioned above, we expect that DL is also practical and effective for DILI prediction.…”

Are the Sublimation Thermodynamics of Organic Molecules Predictable?

“…Many of these solvation/hydration methodologies have been summarized in a recent review by Skyner et al 19 Commonly, QSPR methods are employed due to their speed, convenience and accuracy, when provided with a suitable training dataset. 10,[20][21][22]23 These methods represent the current state-of-the-art in practical solubility prediction. However, QSPR methods lack the theoretical basis of a fundamental physical theory, hence limiting their interpretability and the understanding that can be gained from their use.…”

In silicoADME/T modelling for rational drug design

“…In addition, many 'black box' solubility prediction models were reported in early years. Recently, Lusci et al (2013) reported an aqueous solubility prediction model based on undirected graph recursive neural networks (UG-RNNs) with deep architectures and deep learning. The main advantage of the UG-RNN approach is that it can automatically extract the internal representations from the molecular graphs, as suited for solubility prediction.…”