You are here:
Quantitative Structure--Activity Relationship Modeling of Rat Acute Toxicity by Oral Exposure
Citation:
Zhu, H., T. M. MARTIN, L. Ye, A. Sedykh, D. M. YOUNG, AND A. Tropsha. Quantitative Structure--Activity Relationship Modeling of Rat Acute Toxicity by Oral Exposure. CHEMICAL RESEARCH IN TOXICOLOGY. American Chemical Society, Washington, DC, 22(12):1913-1921, (2009).
Impact/Purpose:
To inform the public
Description:
Background: Few Quantitative Structure-Activity Relationship (QSAR) studies have successfully modeled large, diverse rodent toxicity endpoints. Objective: In this study, a combinatorial QSAR approach has been employed for the creation of robust and predictive models of acute toxicity in rats caused by oral exposure to chemicals. Methods and Results: A comprehensive dataset of 7,385 compounds with their most conservative lethal dose (LD50) values has been compiled. To enable fair comparison between the predictive power of models generated in this study versus a commercial toxicity predictor, TOPKAT (Toxicity Prediction by Komputer Assisted Technology), a modeling subset of the entire dataset was selected that included all compounds used in TOPKAT’s training set; thus, the modeling set was comprised of 3,472 compounds. The remaining 3,913 compounds which were not present in the TOPKAT training set were used as the external validation set. Five different types of QSAR models of rat LD50 were developed for the modeling set. The prediction accuracy for the external validation set was estimated by the coefficient (R2) of linear regression between actual and predicted LD50 values. The use of applicability domain threshold implemented in most models generally improved the external prediction accuracy but obviously led to the decrease in chemical space coverage; thus, depending on the applicability domain threshold R2 ranged from 0.24 to 0.70. Ultimately, several consensus models were developed by averaging the predicted LD50 for every compound using all 5 models. Conclusions: The consensus models afforded higher prediction accuracy for the external validation dataset with the highest coverage as compared to individual constituent models. The validated consensus LD50 models developed in this study can be used as reliable computational predictors of in vivo acute toxicity. The models will be made publicly available from the participating laboratories but we encourage interested investigators send us their compounds for LD50 prediction.
URLs/Downloads:
QUANTITATIVE STRUCTURE--ACTIVITY RELATIONSHIP MODELING OF RAT ACUTE TOXICITY BY ORAL EXPOSURE