Activity landscape of DNA methyltransferase inhibitors bridges chemoinformatics with epigenetic drug discovery

Abstract

Introduction: Activity landscapes are valuable tools for exploring systematically the structure–activity relationships (SAR) of chemical databases. Their application to analyze the SAR of DNA methyltransferase (DNMT) inhibitors, which are attractive compounds as potential epi-drugs or epi-probes, provides useful information to identify pharmacophoric regions and plan the development of predictive models and virtual screening.

Areas covered: This paper highlights different approaches for conducting SAR analysis of datasets with a particular focus on the activity landscape methodology. SAR information of DNMT inhibitors (DNMTi), stored in a public database, is surveyed to further illustrate concepts and generalities of activity landscape modeling with a special emphasis on structure–activity similarity (SAS) maps.

Expert opinion: The increasing SAR information reported for DNMTi opens up avenues to implement activity landscape methods. Despite several activity landscape methods, such as SAS maps, being well established, these need further refinement. For instance, novel combinations of multiple representations, such as the addition of Z-values of similarity (fusion-Z), lead to more robust representations of consensus SAS maps. Density SAS maps improve the visualization of the SAR. A survey of activity cliffs (i.e., pairs of compounds with high structural similarity but high differences in potency) of DNMTi available in a public database suggest that it is feasible to develop predictive models for non-nucleoside DNMTi using approaches such as quantitative structure-activity relationships and that non-nucleoside DNMTi in ChEMBL can be used as query molecules in similarity-based virtual screening.

Keywords:

• activity cliffs
• cancer
• data mining
• DNA methyltransferase
• epi-informatics
• structure-activity relationships

Acknowledgments

The authors thank the assistance of Mariana Giralda Herrera Núñez for obtaining the dataset of DNMT inhibitors. This work is dedicated to the memory of Dr Andoni Garritz Ruiz.

Declaration of interest

The authors are funded in part by the National Autonomous University of Mexico (UNAM), grant number PAIP 5000-9163. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Notes

This box summarizes key points contained in the article.