Development and Application of Computational Methods and Tools for Adverse Drug Reaction and Toxicity Prediction

Abstract

Toxicity is one of the primary reasons for the failure of drug candidates and adverse drug reactions (ADRs) cause significant morbidity and mortality. Therefore, it is important to detect them at the early stage of drug development process. Computational method such as quantitative structure?activity relationship (QSAR) has been explored as complementary method for toxicity prediction. Nevertheless, there are still limitations for current QSAR modeling process which affect the performance and application of QSAR models. This thesis attempts to address these issues with various strategies. Four QSAR models were developed to facilitate the prediction of drug candidates with propensity of various ADRs and toxicities. The methods investigated in this work could improve the quality of QSAR models. Moreover, the QSAR models are potentially useful in drug discovery and clinical practice. Lastly, the software program implemented with integration of peer reviewed models provides an option for users to achieve reliable ADR and toxicity prediction.