Computational Models of Chemical Activity

Using structural data to generate activity predictions for new or poorly characterized chemicals can help researchers and regulators make decisions about further testing needs.

NICEATM provides support for continued development of the Open Structure-activity/property Relationship App (OPERA). OPERA is a free and open-source/open-data suite of QSAR models providing predictions on physicochemical properties, environmental fate and toxicity endpoints.

List of projects for chemical activity computer models
Project	Description	Publication
Open-source quantitative structure-property relationship tools	NICEATM and collaborators at EPA developed tools that use molecular structures to predict the physicochemical features for a wide range of substances.	Zang Q, et al. In silico prediction of physicochemical properties of environmental chemicals using molecular fingerprints and machine learning. J Chem Inf Model. 2017 Jan 23;57(1):36-49.
QSAR models to screen for potential skin sensitizers	NICEATM and collaborators at the University of North Carolina-Chapel Hill (UNC-CH) developed QSAR models of human data that can either be combined with or used instead of animal data to screen for potential skin sensitizers.	Alves VM, et al. QSAR models of human data can enrich or replace LLNA testing for human skin sensitization. Green Chem. 2016 Oct;18:6501-6515.
Open-source tools for predicting estrogen and androgen receptor pathway activity	NICEATM is creating open-source versions of published computational models to predict activity relevant to endocrine disruption. Predictions from these models are available through NICEATM's Integrated Chemical Environment (ICE) resource.	Judson RS, et al. Integrated model of chemical perturbations of a biological pathway using 18 in vitro high-throughput screening assays for the estrogen receptor. Toxicol Sci. 2015 Nov;148(1):137-154. Kleinstreuer NC, et al. Development and validation of a computational model for androgen receptor activity. Chem Res Toxicol. 2017 Apr 17;30(4):946–964.
QSAR models to predict critical parameters for in vitro to in vivo extrapolation (IVIVE)	To support an open-source workflow for IVIVE, NICEATM developed QSAR models to predict properties, such as human plasma fraction unbound and hepatic intrinsic clearance, that affect how substances behave in biological systems.	Property predictions available via OPERA and used in ICE
International collaborations to predict estrogenic, anti-androgenic, and acute systemic toxicity	NICEATM and EPA ran three global collaborations to leverage the expertise of the world-wide modeling community to predict estrogenic activity (CERAPP), anti-androgenic activity (CoMPARA) and acute oral systemic toxicity (CATMoS). Predictions from these models are available through ICE and will be available through the EPA Chemistry Dashboard. The consensus models from CERAPP, CoMPARA, and CATMoS were also added to the standalone OPERA application.	Mansouri K, et al. CERAPP: Collaborative Estrogen Receptor Activity Prediction Project. Environ Health Perspect. 2016 Jul 124:1023–1033. Mansouri K, et al. CoMPARA: Collaborative Modeling Project for Androgen Receptor Activity. Environ Health Perspect. 2020. Kleinstreuer NC, et al. Predictive models for acute oral systemic toxicity: A workshop to bridge the gap from research to regulation. Computational Toxicology. 2018 Nov 8:21–24. Mansouri K, et al. CATMoS: Collaborative Acute Toxicity Modeling Suite. (submitted).
QSAR models to predict physicochemical properties, environmental fate and toxicological endpoints	OPERA was updated to include QSAR models to predict physicochemical properties such as lipid-aqueous dissociation coefficient (logD) and acid dissociation constant (pKa).	Mansouri K, et al. Open-source QSPR models for pKa prediction using multiple machine learning approaches. J Cheminformatics. 2019 11:60.
Statistical models for classification of eye irritants	NICEATM developed statistical models that could potentially be used to classify chemicals as eye corrosives, irritants, or non-irritants according to EPA and GHS hazard classification endpoints. Results suggest that these models are useful for screening substances for eye irritation potential.	Sedykh et al. Manuscript in preparation.
Development of additional QSAR models	NICEATM is developing QSAR models to predict the following: Substrate selectivity for glucuronidation Intestinal cell permeability Acute toxicity endpoints: skin sensitization potency, ocular irritation, dermal irritation, acute systemic inhalation toxicity Fish acute toxicity (LC50) Cancer hallmarks and key characteristics of carcinogens Mechanisms of developmental toxicity	In progress