ToxicR: A computational platform in R for computational toxicology and dose-response analyses

Area of Science:

• Computational toxicology
• Bioinformatics
• Environmental health science

Background:

• High-throughput omics platforms generate complex data requiring advanced computational analysis.
• Existing toxicological software often lacks flexibility for novel research questions.
• A need exists for a stable, open-source platform enabling custom algorithm development in toxicology.

Purpose of the Study:

• To introduce ToxicR, an open-source R package designed to address the limitations of current computational toxicology software.
• To provide a flexible platform for analyzing complex toxicogenomic and other omics data.
• To implement standard analyses used by the National Toxicology Program (NTP) and US Environmental Protection Agency (EPA).

Main Methods:

• Development of ToxicR, an R programming package leveraging existing codebases from EPA's Benchmark Dose software and NTP's BMDExpress.
• Implementation of dose-response analyses for continuous and dichotomous data using Bayesian, maximum likelihood, and model averaging methods.
• Inclusion of standard NTP tests for rodent toxicology and carcinogenicity studies, such as poly-K and Jonckheere trend tests.

Main Results:

• ToxicR provides a flexible platform for computational toxicology analyses, integrating standard NTP and EPA methods.
• The package facilitates custom workflow development for analyzing toxicogenomic data.
• ToxicR's architecture allows for future modular expansion and increased functionality.

Conclusions:

• ToxicR offers a versatile, open-source solution for computational toxicology, enhancing data analysis capabilities.
• The package supports researchers in addressing complex questions in environmental health and toxicology.
• Its flexible design promotes future development and broader application across scientific fields.