Discovering structural alerts for mutagenicity using stable emerging molecular patterns

Abstract : This study is dedicated to an introduction of a novel method that automatically extracts potential structural alerts from a dataset of molecules. These triggering structures can be further used for knowledge discovery and for classification purposes. Computation of the structural alerts results from an implementation of a sophisticated workflow which integrates a graph-mining tool guided by growth-rate and stability. The growth-rate is a well-established measurement of contrast between classes. Moreover, the extracted patterns correspond to formal concepts; the most robust patterns, named the stable emerging patterns (SEPs), can then be identified thanks to their stability, a new notion originating from the domain of Formal Concept Analysis. All these elements are explained in the paper from the point of view of computation. The method was applied on a molecular dataset on mutagenicity. The experimental results demonstrate its efficiency: it automatically outputs a manageable amount of structural patterns that are strongly related to mutagenicity. Moreover, a part of the resulting structures corresponds to already known structural alerts. Finally, an in-depth chemical analysis relies on these structures, it demonstrates how the method can initiate promising processes of chemical knowledge discovery.
Document type :
Journal articles
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01186716
Contributor : Aleksey Buzmakov <>
Submitted on : Tuesday, August 25, 2015 - 3:19:32 PM
Last modification on : Tuesday, July 23, 2019 - 2:54:03 PM

Identifiers

Citation

Jean-Philippe Metivier, Alban Lepailleur, Aleksey Buzmakov, Guillaume Poezevara, Bruno Crémilleux, et al.. Discovering structural alerts for mutagenicity using stable emerging molecular patterns. Journal of Chemical Information and Modeling, American Chemical Society, 2015, 55 (5), pp.925--940. ⟨10.1021/ci500611v⟩. ⟨hal-01186716⟩

Share

Metrics

Record views

535