Quantitative Structure-Activity Relationship Models for Predicting Risk of Drug-Induced Liver Injury in Humans

Document Type

Article

Publication Date

3-21-2018

Department

Computing

School

Computing Sciences and Computer Engineering

Abstract

Drug-induced liver injury (DILI) risk in humans is a complicated safety concern due to diverse mechanisms, various severity levels, variation in population groups, and difficulty in annotation of drugs, especially for the drugs that have been on the market for a short period of time. DILI remains a challenge for the industry and regulatory agencies. Assessing DILI risk in humans is important to assist drug development for the industry and to inform decision making on safety evaluation of drug products in regulatory science. Though various experimental methods have been used in current practices for assessment of DILI risk, in silico methods have been adopted in the field as an alternative because the development and validation of in silico models are much faster and cheaper. Many quantitative structure–activity relationship (QSAR) DILI prediction models have been reported. To better understand the QSAR models reported and to foster development of more reliable QSAR models, this chapter provides an instruction to the principals and the components of QSAR modeling, a summary on some popular algorithms and tools for QSAR modeling, and a review of QSAR models developed for prediction of DILI.

Publication Title

Drug Induced Liver Toxicity

First Page

77

Last Page

100

Share

COinS