CORAL: Binary classifications (active/inactive) for drug-induced liver injury

Alla P Toropova, Andrey A Toropov

Research output: Contribution to journalArticle

Abstract

INTRODUCTION: The data on human hepatotoxcity (drug-induced liver injury) is extremely important information from point of view of drug discovery. Experimental clinical data on this endpoint is scarce. Experimental way to extend databases on this endpoint is extremely difficult. Quantitative structure - activity relationships (QSAR) is attractive alternative of the experimental approach.

METHODS: Predictive models for human hepatotoxicity (drug-induced liver injury) have been built up by the Monte Carlo method with using of the CORAL software (http://www.insilico.eu/coral). These models are the binary classifications into active class and inactive class. These models are calculated with so-called "semi correlations" described in this work. The Mattews correlation coefficient of these models for external validation sets ranged from 0.52 to 0.62.

RESULTS DISCUSSION: The approach has been checked up with a group of random splits into the training and validation sets. These stochastic experiments have shown the stability of results: predictability of the models for various splits. Thus, the attempt to build up the classification QSAR model by means of the Monte Carlo technique, based on representation of the molecular structure via simplified molecular input line entry systems (SMILES) and hydrogen suppressed graph (HSG) using the CORAL software (http://www.insilico.eu/coral) has shown ability of this approach to provide quite good prediction of the examined endpoint (drug-induced liver injury).

Original languageEnglish
Pages (from-to)51-57
Number of pages7
JournalToxicology Letters
Volume268
DOIs
Publication statusPublished - Feb 15 2017

Keywords

  • Chemical and Drug Induced Liver Injury
  • Computer Simulation
  • Drug Discovery
  • Humans
  • Models, Biological
  • Models, Statistical
  • Molecular Structure
  • Monte Carlo Method
  • Quantitative Structure-Activity Relationship
  • Reproducibility of Results
  • Risk Assessment
  • Risk Factors
  • Software
  • Xenobiotics
  • Journal Article

Fingerprint Dive into the research topics of 'CORAL: Binary classifications (active/inactive) for drug-induced liver injury'. Together they form a unique fingerprint.

  • Cite this