S3: Modeling the Tox21 10K chemical profiles for in vivo toxicity prediction

Modeling the Tox21 10K chemical profiles for in vivo toxicity prediction and adverse outcome pathway characterization

S3: Modeling the Tox21 10K chemical profiles for in vivo toxicity prediction, OpenTox USA 2017
PRESENTING AUTHOR: 

Ruili Huang

INSTITUTION / COMPANY : 

NIH

POSITION: 

Lead Informatics Scientist

ABSTRACT CONTENT / DETAILS: 

Target-specific, mechanism-oriented in vitro assays pose a promising alternative to traditional animal toxicology studies. The Tox21 program, a large-scale in vitro chemical toxicity screening effort, has tested approximately 10,000 chemicals in triplicates at 15 concentrations against a panel of nuclear receptor and stress response pathway assays, producing more than 70 million data points to date that form a rich set of compound in vitro activity profiles. Clustering analyses of these activity profiles across the assays reveal relationships useful for the generation of hypotheses on compound mode of action (MOA). Compound activity data are applied in combination with structural information to build predictive models for a number of in vivo toxicity endpoints. The in vitro assay data based models perform better in predicting human toxicity endpoints than animal toxicity, while combining structure and activity data results in better models than using structure or activity data alone. Pathways represented by assays that contribute the most to in vivo toxicity predictions suggest potential adverse outcome pathways (AOPs). These results show that in vitro activity profiles can be applied as signatures of compound mechanism of toxicity that provide valuable information for the development of AOPs.