Abstract:A bstract Among virtual screening methods that have been developed to facilitate the drug discovery process, chemogenomics presents the particularity to tackle the question of predicting ligands for proteins, at at scales both in the protein and chemical spaces. Therefore, in addition to to predict drug candidates for a given therapeutic protein target, like more classical ligand-based or receptor-based methods do, chemogenomics can also predict off-targets at the proteome level, and therefore, identify potential side-effects or drug repositioning opportunities. In this study, we study and compare machine-learning and deep learning approaches for chemogenomics, that are applicable to screen large sets of compounds against large sets of druggable proteins. State-of-the-art drug chemogenomics methods rely on expert-based chemical and protein descriptors or similarity measures. The recent development of deep learning approaches enabled to design algorithms that learn numerical abstract representations of molecular graphs and protein sequences in an end-to-end fashion, i.e., so that the learnt features optimise the objective function of the drug-target interaction prediction task. In this paper, we address drug-target interaction prediction at the druggable proteome-level, with what we define as the chemogenomic neuron network. This network consists of a feed-forward neuron network taking as input the combination of molecular and protein representations learnt by molecular graph and protein sequence encoders. We first propose a standard formulation of this chemogenomic neuron network. Then, we compare the performances of the standard chemogenomic network to reference deep learning or shallow (machine-learning without deep learning) methods. In particular, we show that such a representation learning approach is competitive with state-of-the-art chemogenomics with shallow methods, but not ultimately superior. We evaluate the most promising neuron network architectures and data augmentation techniques, such as multi-view and transfer learning, to improve the prediction performance of the chemogenomic network. Our results shed new insights on the design of chemogenomics approaches based on representation learning algorithms. Most importantly, we conclude from our observations that a promising research direction is to integrate heterogeneous sources of data such as various bioactivity datasets, or independently, multiple molecule and protein attribute views, instead of focusing on sophisticated, yet intuitively relevant, encoder’s neuron network architecture.

Chemception: A Deep Neural Network with Minimal Chemistry Knowledge Matches the Performance of Expert-developed QSAR/QSPR Models

Deep learning for computational chemistry

A Deep Learning-Based Chemical System for QSAR Prediction

Chemi-net: a graph convolutional network for accurate drug property prediction

Deep Learning for Chemoinformatics

Molecular Graph Encoding Convolutional Neural Networks for Automatic Chemical Feature Extraction.

Quantum-chemical insights from deep tensor neural networks

Evaluation of deep and shallow learning methods in chemogenomics for the prediction of drugs specificity

Evaluation of network architecture and data augmentation methods for deep learning in chemogenomics

Prediction of chemical compounds properties using a deep learning model

CheMixNet: Mixed DNN Architectures for Predicting Chemical Properties using Multiple Molecular Representations

Accurate predictions of drugs aqueous solubility via deep learning tools

Geometric deep learning for molecular property predictions with chemical accuracy across chemical space

Deep Learning and Computational Chemistry

Accurate Predictions of Aqueous Solubility of Drug Moleculesviathe Multilevel Graph Convolutional Network (MGCN) and SchNet Architectures

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

An Architecture of Deep Learning in QSPR Modeling for the Prediction of Critical Properties Using Molecular Signatures

Spectroscopy-Guided Deep Learning Predicts Solid-Liquid Surface Adsorbate Properties in Unseen Solvents.

ElemNet: Deep Learning the Chemistry of Materials From Only Elemental Composition

Quantum-Chemical Insights from Interpretable Atomistic Neural Networks