Toxicophore exploration as a screening technology for drug design and discovery: techniques, scope and limitations

Abstract

Toxicity is a common drawback of newly designed chemotherapeutic agents. With the exception of pharmacophore-induced toxicity (lack of selectivity at higher concentrations of a drug), the toxicity due to chemotherapeutic agents is based on the toxicophore moiety present in the drug. To date, methodologies implemented to determine toxicophores may be broadly classified into biological, bioanalytical and computational approaches. The biological approach involves analysis of bioactivated metabolites, whereas the computational approach involves a QSAR-based method, mapping techniques, an inverse docking technique and a few toxicophore identification/estimation tools. Being one of the major steps in drug discovery process, toxicophore identification has proven to be an essential screening step in drug design and development. The paper is first of its kind, attempting to cover and compare different methodologies employed in predicting and determining toxicophores with an emphasis on their scope and limitations. Such information may prove vital in the appropriate selection of methodology and can be used as screening technology by researchers to discover the toxicophoric potentials of their designed and synthesized moieties. Additionally, it can be utilized in the manipulation of molecules containing toxicophores in such a manner that their toxicities might be eliminated or removed.

DeepSynergy: predicting anti-cancer drug synergy with Deep Learning

“…The set of chemical features was completed by binary toxicophore features based on a set of toxicophores collected from the literature. Toxicophores are substructures which are known to be toxic ( Singh et al , 2016 ). The chemical feature space was reduced by filtering out zero variance features.…”

Managing the challenge of drug-induced liver injury: a roadmap for the development and deployment of preclinical predictive models

“…The links between compound characteristics of interest (e.g., physicochemical properties, bioactivation, and general toxicity) and molecular structure (i.e., 'the similarity principle') form the basis of these in silico applications that differ only in terms of complexity and performance 43 . For example, various toxicophores or problematic substructures can be identified with this approach 44,45 . These computational assessments are typically conducted prior to TIER I assays in order to prevent the synthesis of compounds with low probability of success or to prioritize screening of subsets of compounds, thereby optimizing drug discovery efforts.…”

Elucidating the aryl hydrocarbon receptor antagonism from a chemical-structural perspective

“…Toxicophoric modelling aims to identify a set of features or restrictions in the molecules and/or their receptors that determine their capacity to interact in any mode and consequently cause toxic effects. These models may be built following structure-based or ligand-based approaches depending on whether the information of the target receptor is employed or not [48,49].…”