Identification of a Preferred Set of Molecular Descriptors for Compound Classification Based on Principal Component Analysis Export

@article{citeulike:1470853, abstract = {An algorithm based on principal component analysis was investigated to classify molecules in a database consisting of 455 compounds with activities against seven different biological targets. Diversity profiles of these compound sets were calculated and compared. To effectively classify compounds with similar biological activity, all possible combinations of 17 molecular descriptors were tested by complete factorial analysis, and preferred descriptor combinations were identified. High efficiency was achieved for a combination of a limited set of structural keys and two or three additional 2D descriptors. The performance of the approach was compared to JarvisPatrick clustering.}, address = {Computational Chemistry and Informatics, MDS Panlabs, 11804 North Creek Parkway, Bothell, Washington 98011, and MDS Panlabs and Department of Biological Structure, University of Washington, Seattle, Washington 98195}, author = {Xue, Ling and Godden, Jeff and Gao, Hua and Bajorath, Jurgen}, citeulike-article-id = {1470853}, citeulike-linkout-0 = {http://dx.doi.org/10.1021/ci980231d}, citeulike-linkout-1 = {http://pubs.acs.org/doi/abs/10.1021/ci980231d}, day = {1}, doi = {10.1021/ci980231d}, journal = {Journal of Chemical Information and Computer Sciences}, keywords = {compound-classification, feature-selection, fingerprinting, pca}, month = {July}, number = {4}, pages = {699--704}, posted-at = {2009-10-19 18:53:47}, priority = {2}, title = {Identification of a Preferred Set of Molecular Descriptors for Compound Classification Based on Principal Component Analysis}, url = {http://dx.doi.org/10.1021/ci980231d}, volume = {39}, year = {1999} }

TY - JOUR ID - citeulike:1470853 L3 - citeulike-article-id:1470853 N2 - An algorithm based on principal component analysis was investigated to classify molecules in a database consisting of 455 compounds with activities against seven different biological targets. Diversity profiles of these compound sets were calculated and compared. To effectively classify compounds with similar biological activity, all possible combinations of 17 molecular descriptors were tested by complete factorial analysis, and preferred descriptor combinations were identified. High efficiency was achieved for a combination of a limited set of structural keys and two or three additional 2D descriptors. The performance of the approach was compared to JarvisPatrick clustering. IS - 4 JF - Journal of Chemical Information and Computer Sciences AD - Computational Chemistry and Informatics, MDS Panlabs, 11804 North Creek Parkway, Bothell, Washington 98011, and MDS Panlabs and Department of Biological Structure, University of Washington, Seattle, Washington 98195 EP - 704 TI - Identification of a Preferred Set of Molecular Descriptors for Compound Classification Based on Principal Component Analysis VL - 39 SP - 699 KW - compound-classification KW - feature-selection KW - fingerprinting KW - pca AU - Xue, Ling AU - Godden, Jeff AU - Gao, Hua AU - Bajorath, Jurgen PY - 1999/07/01/ UR - http://dx.doi.org/10.1021/ci980231d ER -