ToxicBlend: Virtual Screening of Toxic Compounds with Ensemble Predictors

06/12/2018

∙

Timely assessment of compound toxicity is one of the biggest challenges facing the pharmaceutical industry today. A significant proportion of compounds identified as potential leads are ultimately discarded due to the toxicity they induce. In this paper, we propose a novel machine learning approach for the prediction of molecular activity on ToxCast targets. We combine extreme gradient boosting with fully-connected and graph-convolutional neural network architectures trained on QSAR physical molecular property descriptors, PubChem molecular fingerprints, and SMILES sequences. Our ensemble predictor leverages the strengths of each individual technique, significantly outperforming existing state-of-the art models on the ToxCast and Tox21 toxicity-prediction datasets. We provide free access to molecule toxicity prediction using our model at http://www.owkin.com/toxicblend.

READ FULL TEXT

ToxicBlend: Virtual Screening of Toxic Compounds with Ensemble Predictors

Virtual Screening of Pharmaceutical Compounds with hERG Inhibitory Activity (Cardiotoxicity) using Ensemble Learning

Are Learned Molecular Representations Ready For Prime Time?

An ensemble of VisNet, Transformer-M, and pretraining models for molecular property prediction in OGB Large-Scale Challenge @ NeurIPS 2022

Molecular activity prediction using graph convolutional deep neural network considering distance on a molecular graph

Gode – Integrating Biochemical Knowledge Graph into Pre-training Molecule Graph Neural Network

Adaptive Invariance for Molecule Property Prediction

ALMERIA: Boosting pairwise molecular contrasts with scalable methods

ToxicBlend: Virtual Screening of Toxic Compounds with Ensemble Predictors

Related Research

Virtual Screening of Pharmaceutical Compounds with hERG Inhibitory Activity (Cardiotoxicity) using Ensemble Learning

Are Learned Molecular Representations Ready For Prime Time?

An ensemble of VisNet, Transformer-M, and pretraining models for molecular property prediction in OGB Large-Scale Challenge @ NeurIPS 2022

Molecular activity prediction using graph convolutional deep neural network considering distance on a molecular graph

Gode – Integrating Biochemical Knowledge Graph into Pre-training Molecule Graph Neural Network

Adaptive Invariance for Molecule Property Prediction

ALMERIA: Boosting pairwise molecular contrasts with scalable methods