Share This:

Computer Models of Chemical Activity

Using structural data to generate activity predictions for new or poorly characterized chemicals can inform decisions about further testing needs.

Project Description Publication
Open-source quantitative structure-property relationship tools NICEATM and collaborators at EPA developed tools that use molecular structures to estimate the physicochemical features for a wide range of chemicals. Zang et al. 2017. In silico prediction of physicochemical properties of environmental chemicals using molecular fingerprints and machine learning. J Chem Inf Modeling 57:36-49
Quantitative structure–activity relationship models to screen for potential skin sensitizers Working with collaborators at the University of North Carolina-Chapel Hill, NICEATM scientists developed QSAR models of human data that can be combined with or used instead of animal data to screen for potential skin sensitizers. Alves et al. 2016. QSAR models of human data can enrich or replace LLNA testing for human skin sensitization. Green Chemistry 18:6501-6515
Domain-specific QSAR models for phenolic compounds NICEATM and UNC-CH collaborators also developed domain-specific QSAR models to predict specific activity and relative potency of phenolic compounds that act on the estrogen receptor pathway. In progress
Open source tools for predicting estrogen and androgen receptor activity NICEATM is creating open-source versions of published and unpublished QSAR models to predict chemical activity relevant to endocrine disruption. Predictions from these models will be available via NICEATM’s Integrated Chemical Environment resource. In progress
QSAR models to predict critical parameters for in vitro to in vivo extrapolation As part of an open-source workflow for IVIVE, NICEATM is developing QSAR models to predict Henry’s constant, dissociation constant, and tissue-specific partition coefficients. In progress