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

Abstract

In the last few years, we have seen the transformative impact of deep learning in many applications, particularly in speech recognition and computer vision. Inspired by Google's Inception-ResNet deep convolutional neural network (CNN) for image classification, we have developed "Chemception", a deep CNN for the prediction of chemical properties, using just the images of 2D drawings of molecules. We develop Chemception without providing any additional explicit chemistry knowledge, such as basic concepts like periodicity, or advanced features like molecular descriptors and fingerprints. We then show how Chemception can serve as a general-purpose neural network architecture for predicting toxicity, activity, and solvation properties when trained on a modest database of 600 to 40,000 compounds. When compared to multi-layer perceptron (MLP) deep neural networks trained with ECFP fingerprints, Chemception slightly outperforms in activity and solvation prediction and slightly underperforms in toxicity prediction. Having matched the performance of expert-developed QSAR/QSPR deep learning models, our work demonstrates the plausibility of using deep neural networks to assist in computational chemistry research, where the feature engineering process is performed primarily by a deep learning algorithm.

Submitted to a chemistry peer-reviewed journal

References in corpus (8)

• Deep Learning in Neural Networks: An Overview
• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Going Deeper with Convolutions
• cuDNN: Efficient Primitives for Deep Learning
• Massively Multitask Networks for Drug Discovery
• Multi-task Neural Networks for QSAR Predictions
• MoleculeNet: A Benchmark for Molecular Machine Learning

Cited by in corpus (21)

• Deep learning for molecular design - a review of the state of the art
• Autonomous discovery in the chemical sciences part II: Outlook
• Exploring Chemical Space using Natural Language Processing Methodologies for Drug Discovery
• Using Attribution to Decode Dataset Bias in Neural Network Models for Chemistry
• Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design -- Toward a Unified Approach: State-of-the-Art and Future Directions
• Machine learning and AI-based approaches for bioactive ligand discovery and GPCR-ligand recognition
• Molecule Attention Transformer
• ChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning
• Learning physical properties of liquid crystals with deep convolutional neural networks
• SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties
• Reliable Prediction Errors for Deep Neural Networks Using Test-Time Dropout
• Using Molecular Embeddings in QSAR Modeling: Does it Make a Difference?
• CheMixNet: Mixed DNN Architectures for Predicting Chemical Properties using Multiple Molecular Representations
• Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
• Multimodal Deep Neural Networks using Both Engineered and Learned Representations for Biodegradability Prediction
• Artificial Intelligence in Drug Discovery: Applications and Techniques
• A Systematic Approach to Featurization for Cancer Drug Sensitivity Predictions with Deep Learning
• End-to-End Attention-based Image Captioning
• Augmenting Molecular Images with Vector Representations as a Featurization Technique for Drug Classification
• IL-Net: Using Expert Knowledge to Guide the Design of Furcated Neural Networks
• Generate Novel Molecules With Target Properties Using Conditional Generative Models