POIMs: positional oligomer importance matrices--understanding support vector machine-based signal detectors Export

@article{citeulike:2966378, abstract = {Motivation: At the heart of many important bioinformatics problems, such as gene finding and function prediction, is the classification of biological sequences. Frequently the most accurate classifiers are obtained by training support vector machines (SVMs) with complex sequence kernels. However, a cumbersome shortcoming of SVMs is that their learned decision rules are very hard to understand for humans and cannot easily be related to biological facts. Results: To make SVM-based sequence classifiers more accessible and profitable, we introduce the concept of positional oligomer importance matrices (POIMs) and propose an efficient algorithm for their computation. In contrast to the raw SVM feature weighting, POIMs take the underlying correlation structure of k-mer features induced by overlaps of related k-mers into account. POIMs can be seen as a powerful generalization of sequence logos: they allow to capture and visualize sequence patterns that are relevant for the investigated biological phenomena. Availability: All source code, datasets, tables and figures are available at http://www.fml.tuebingen.mpg.de/raetsch/projects/POIM. Contact: Soeren.Sonnenburg@first.fraunhofer.de Supplementary information: Supplementary data are available at Bioinformatics online. 10.1093/bioinformatics/btn170}, author = {Sonnenburg, Soren and Zien, Alexander and Philips, Petra and Ratsch, Gunnar}, booktitle = {Bioinformatics}, citeulike-article-id = {2966378}, citeulike-linkout-0 = {http://dx.doi.org/10.1093/bioinformatics/btn170}, citeulike-linkout-1 = {http://bioinformatics.oxfordjournals.org/cgi/content/abstract/24/13/i6}, citeulike-linkout-2 = {http://view.ncbi.nlm.nih.gov/pubmed/18586746}, citeulike-linkout-3 = {http://www.hubmed.org/display.cgi?uids=18586746}, day = {1}, doi = {10.1093/bioinformatics/btn170}, journal = {Bioinformatics}, keywords = {k-mer}, month = {July}, number = {13}, pages = {i6--14}, posted-at = {2008-08-10 18:48:00}, priority = {3}, title = {POIMs: positional oligomer importance matrices--understanding support vector machine-based signal detectors}, url = {http://dx.doi.org/10.1093/bioinformatics/btn170}, volume = {24}, year = {2008} }

TY - JOUR ID - citeulike:2966378 L3 - citeulike-article-id:2966378 N2 - Motivation: At the heart of many important bioinformatics problems, such as gene finding and function prediction, is the classification of biological sequences. Frequently the most accurate classifiers are obtained by training support vector machines (SVMs) with complex sequence kernels. However, a cumbersome shortcoming of SVMs is that their learned decision rules are very hard to understand for humans and cannot easily be related to biological facts. Results: To make SVM-based sequence classifiers more accessible and profitable, we introduce the concept of positional oligomer importance matrices (POIMs) and propose an efficient algorithm for their computation. In contrast to the raw SVM feature weighting, POIMs take the underlying correlation structure of k-mer features induced by overlaps of related k-mers into account. POIMs can be seen as a powerful generalization of sequence logos: they allow to capture and visualize sequence patterns that are relevant for the investigated biological phenomena. Availability: All source code, datasets, tables and figures are available at http://www.fml.tuebingen.mpg.de/raetsch/projects/POIM. Contact: Soeren.Sonnenburg@first.fraunhofer.de Supplementary information: Supplementary data are available at Bioinformatics online. 10.1093/bioinformatics/btn170 IS - 13 JF - Bioinformatics EP - 14 TI - POIMs: positional oligomer importance matrices--understanding support vector machine-based signal detectors VL - 24 T2 - Bioinformatics SP - i6 KW - k-mer AU - Sonnenburg, Soren AU - Zien, Alexander AU - Philips, Petra AU - Ratsch, Gunnar PY - 2008/07/01/ UR - http://dx.doi.org/10.1093/bioinformatics/btn170 ER -