Discriminative Chemical Patterns: Automatic and Interactive Design. Bietz, S., Schomburg, K. T., Hilbig, M., & Rarey, M. Journal of Chemical Information and Modeling, 55(8):1535–1546, August, 2015. Publisher: American Chemical Society
Discriminative Chemical Patterns: Automatic and Interactive Design [link]Paper  doi  abstract   bibtex   
The classification of molecules with respect to their inhibiting, activating, or toxicological potential constitutes a central aspect in the field of cheminformatics. Often, a discriminative feature is needed to distinguish two different molecule sets. Besides physicochemical properties, substructures and chemical patterns belong to the descriptors most frequently applied for this purpose. As a commonly used example of this descriptor class, SMARTS strings represent a powerful concept for the representation and processing of abstract chemical patterns. While their usage facilitates a convenient way to apply previously derived classification rules on new molecule sets, the manual generation of useful SMARTS patterns remains a complex and time-consuming process. Here, we introduce SMARTSminer, a new algorithm for the automatic derivation of discriminative SMARTS patterns from preclassified molecule sets. Based on a specially adapted subgraph mining algorithm, SMARTSminer identifies structural features that are frequent in only one of the given molecule classes. In comparison to elemental substructures, it also supports the consideration of general and specific SMARTS features. Furthermore, SMARTSminer is integrated into an interactive pattern editor named SMARTSeditor. This allows for an intuitive visualization on the basis of the SMARTSviewer concept as well as interactive adaption and further improvement of the generated patterns. Additionally, a new molecular matching feature provides an immediate feedback on a pattern’s matching behavior across the molecule sets. We demonstrate the utility of the SMARTSminer functionality and its integration into the SMARTSeditor software in several different classification scenarios.
@article{bietz_discriminative_2015,
	title = {Discriminative {Chemical} {Patterns}: {Automatic} and {Interactive} {Design}},
	volume = {55},
	issn = {1549-9596},
	shorttitle = {Discriminative {Chemical} {Patterns}},
	url = {https://doi.org/10.1021/acs.jcim.5b00323},
	doi = {10.1021/acs.jcim.5b00323},
	abstract = {The classification of molecules with respect to their inhibiting, activating, or toxicological potential constitutes a central aspect in the field of cheminformatics. Often, a discriminative feature is needed to distinguish two different molecule sets. Besides physicochemical properties, substructures and chemical patterns belong to the descriptors most frequently applied for this purpose. As a commonly used example of this descriptor class, SMARTS strings represent a powerful concept for the representation and processing of abstract chemical patterns. While their usage facilitates a convenient way to apply previously derived classification rules on new molecule sets, the manual generation of useful SMARTS patterns remains a complex and time-consuming process. Here, we introduce SMARTSminer, a new algorithm for the automatic derivation of discriminative SMARTS patterns from preclassified molecule sets. Based on a specially adapted subgraph mining algorithm, SMARTSminer identifies structural features that are frequent in only one of the given molecule classes. In comparison to elemental substructures, it also supports the consideration of general and specific SMARTS features. Furthermore, SMARTSminer is integrated into an interactive pattern editor named SMARTSeditor. This allows for an intuitive visualization on the basis of the SMARTSviewer concept as well as interactive adaption and further improvement of the generated patterns. Additionally, a new molecular matching feature provides an immediate feedback on a pattern’s matching behavior across the molecule sets. We demonstrate the utility of the SMARTSminer functionality and its integration into the SMARTSeditor software in several different classification scenarios.},
	number = {8},
	urldate = {2023-08-31},
	journal = {Journal of Chemical Information and Modeling},
	author = {Bietz, Stefan and Schomburg, Karen T. and Hilbig, Matthias and Rarey, Matthias},
	month = aug,
	year = {2015},
	note = {Publisher: American Chemical Society},
	pages = {1535--1546},
	file = {ACS Full Text Snapshot:/Users/flodje_uds/Zotero/storage/ZHQMFACM/acs.jcim.html:text/html},
}

Downloads: 0