Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation. Geppert, H., Vogt, M., & Bajorath, J. J Chem Inf Model, 50(2):205–216, 2010. doi abstract bibtex Data mining methods play a major role in chemoinformatics. In this Perspective, we focus on data mining methodologies that are particularly relevant for ligand-based virtual screening. In order to provide an up-to-date view of the field, we largely concentrate on publications of the past two to three years. We discuss alternative chemical space representations for compound screening, popular data mining methods, and new algorithms. Furthermore, recently developed molecular fingerprints and specialized similarity measures are reviewed. As increasing amounts of public-domain compound bioactivity data become available, data mining approaches are also utilized for new types of virtual screening applications, for example, to search for target-selective molecules or ligands of orphan targets. Moreover, how to best evaluate and compare the performance of different computational screening methodologies has emerged as one of the central questions in chemical data mining. Therefore, method evaluation strategies and approaches for the design of advanced benchmark data sets are also described. Taken together, the survey presented herein makes it possible to highlight a number of trends that can currently be observed in the chemoinformatics field. Furthermore, we also discuss problems associated with conventional benchmark settings for the evaluation of virtual screening methods and the need for community-wide standards.
@Article{geppert10current,
author = {Geppert, Hanna and Vogt, Martin and Bajorath, J{\"{u}}rgen},
title = {Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation},
journal = {J Chem Inf Model},
year = {2010},
volume = {50},
number = {2},
pages = {205--216},
abstract = {Data mining methods play a major role in chemoinformatics. In this Perspective, we focus on data mining methodologies that are particularly relevant for ligand-based virtual screening. In order to provide an up-to-date view of the field, we largely concentrate on publications of the past two to three years. We discuss alternative chemical space representations for compound screening, popular data mining methods, and new algorithms. Furthermore, recently developed molecular fingerprints and specialized similarity measures are reviewed. As increasing amounts of public-domain compound bioactivity data become available, data mining approaches are also utilized for new types of virtual screening applications, for example, to search for target-selective molecules or ligands of orphan targets. Moreover, how to best evaluate and compare the performance of different computational screening methodologies has emerged as one of the central questions in chemical data mining. Therefore, method evaluation strategies and approaches for the design of advanced benchmark data sets are also described. Taken together, the survey presented herein makes it possible to highlight a number of trends that can currently be observed in the chemoinformatics field. Furthermore, we also discuss problems associated with conventional benchmark settings for the evaluation of virtual screening methods and the need for community-wide standards.},
doi = {10.1021/ci900419k},
keywords = {virtual screening; fingerprints;},
owner = {Sebastian},
pmid = {20088575},
timestamp = {2015.03.31},
}
Downloads: 0
{"_id":"qtZzgSRzjko9zGjDE","bibbaseid":"geppert-vogt-bajorath-currenttrendsinligandbasedvirtualscreeningmolecularrepresentationsdataminingmethodsnewapplicationareasandperformanceevaluation-2010","authorIDs":[],"author_short":["Geppert, H.","Vogt, M.","Bajorath, J."],"bibdata":{"bibtype":"article","type":"article","author":[{"propositions":[],"lastnames":["Geppert"],"firstnames":["Hanna"],"suffixes":[]},{"propositions":[],"lastnames":["Vogt"],"firstnames":["Martin"],"suffixes":[]},{"propositions":[],"lastnames":["Bajorath"],"firstnames":["Jürgen"],"suffixes":[]}],"title":"Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation","journal":"J Chem Inf Model","year":"2010","volume":"50","number":"2","pages":"205–216","abstract":"Data mining methods play a major role in chemoinformatics. In this Perspective, we focus on data mining methodologies that are particularly relevant for ligand-based virtual screening. In order to provide an up-to-date view of the field, we largely concentrate on publications of the past two to three years. We discuss alternative chemical space representations for compound screening, popular data mining methods, and new algorithms. Furthermore, recently developed molecular fingerprints and specialized similarity measures are reviewed. As increasing amounts of public-domain compound bioactivity data become available, data mining approaches are also utilized for new types of virtual screening applications, for example, to search for target-selective molecules or ligands of orphan targets. Moreover, how to best evaluate and compare the performance of different computational screening methodologies has emerged as one of the central questions in chemical data mining. Therefore, method evaluation strategies and approaches for the design of advanced benchmark data sets are also described. Taken together, the survey presented herein makes it possible to highlight a number of trends that can currently be observed in the chemoinformatics field. Furthermore, we also discuss problems associated with conventional benchmark settings for the evaluation of virtual screening methods and the need for community-wide standards.","doi":"10.1021/ci900419k","keywords":"virtual screening; fingerprints;","owner":"Sebastian","pmid":"20088575","timestamp":"2015.03.31","bibtex":"@Article{geppert10current,\n author = {Geppert, Hanna and Vogt, Martin and Bajorath, J{\\\"{u}}rgen},\n title = {Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation},\n journal = {J Chem Inf Model},\n year = {2010},\n volume = {50},\n number = {2},\n pages = {205--216},\n abstract = {Data mining methods play a major role in chemoinformatics. In this Perspective, we focus on data mining methodologies that are particularly relevant for ligand-based virtual screening. In order to provide an up-to-date view of the field, we largely concentrate on publications of the past two to three years. We discuss alternative chemical space representations for compound screening, popular data mining methods, and new algorithms. Furthermore, recently developed molecular fingerprints and specialized similarity measures are reviewed. As increasing amounts of public-domain compound bioactivity data become available, data mining approaches are also utilized for new types of virtual screening applications, for example, to search for target-selective molecules or ligands of orphan targets. Moreover, how to best evaluate and compare the performance of different computational screening methodologies has emerged as one of the central questions in chemical data mining. Therefore, method evaluation strategies and approaches for the design of advanced benchmark data sets are also described. Taken together, the survey presented herein makes it possible to highlight a number of trends that can currently be observed in the chemoinformatics field. Furthermore, we also discuss problems associated with conventional benchmark settings for the evaluation of virtual screening methods and the need for community-wide standards.},\n doi = {10.1021/ci900419k},\n keywords = {virtual screening; fingerprints;},\n owner = {Sebastian},\n pmid = {20088575},\n timestamp = {2015.03.31},\n}\n\n","author_short":["Geppert, H.","Vogt, M.","Bajorath, J."],"key":"geppert10current","id":"geppert10current","bibbaseid":"geppert-vogt-bajorath-currenttrendsinligandbasedvirtualscreeningmolecularrepresentationsdataminingmethodsnewapplicationareasandperformanceevaluation-2010","role":"author","urls":{},"keyword":["virtual screening; fingerprints;"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://git.bio.informatik.uni-jena.de/fleisch/literature/raw/master/group-literature.bib","creationDate":"2019-11-19T16:50:41.970Z","downloads":0,"keywords":["virtual screening; fingerprints;"],"search_terms":["current","trends","ligand","based","virtual","screening","molecular","representations","data","mining","methods","new","application","areas","performance","evaluation","geppert","vogt","bajorath"],"title":"Current trends in ligand-based virtual screening: molecular representations, data mining methods, new application areas, and performance evaluation","year":2010,"dataSources":["C5FtkvWWggFfMJTFX"]}