- title = "Motif kernel generated by genetic programming improves remote homology and fold detection",
- author = "Tony Handstad and Arne J H Hestnes and Pal Saetrom",
- journal = "BMC Bioinformatics",
- year = "2007",
- volume = "8",
- number = "23",
- month = jan # "~25",
- publisher = "BioMed Central Ltd.",
- bibsource = "OAI-PMH server at www.biomedcentral.com",
- language = "en",
- oai = "oai:biomedcentral.com:1471-2105-8-23",
- rights = "Copyright 2007 H{\aa}ndstad et al; licensee BioMed Central Ltd.",
- keywords = "genetic algorithms, genetic programming, GPkernel, SVM, MISD, boosting",
- ISSN = "1471-2105",
-
URL = "
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.276.5386",
-
URL = "
http://www.biomedcentral.com/content/pdf/1471-2105-8-23.pdf",
-
URL = "http://www.biomedcentral.com/1471-2105/8/23",
-
DOI = "
doi:10.1186/1471-2105-8-23",
- broken_undergraduate_thesis = "http://www.diva-portal.org/diva/getDocument?urn_nbn_no_ntnu_diva-1030-1__fulltext.pdf",
- size = "16 pages",
-
abstract = "Background
Protein remote homology detection is a central problem in computational biology. Most recent methods train support vector machines to discriminate between related and unrelated sequences and these studies have introduced several types of kernels. One successful approach is to base a kernel on shared occurrences of discrete sequence motifs. Still, many protein sequences fail to be classified correctly for a lack of a suitable set of motifs for these sequences. Results
We introduce the GPkernel, which is a motif kernel based on discrete sequence motifs where the motifs are evolved using genetic programming. All proteins can be grouped according to evolutionary relations and structure, and the method uses this inherent structure to create groups of motifs that discriminate between different families of evolutionary origin. When tested on two SCOP benchmarks, the superfamily and fold recognition problems, the GPkernel gives significantly better results compared to related methods of remote homology detection. Conclusion
The GPkernel gives particularly good results on the more difficult fold recognition problem compared to the other methods. This is mainly because the method creates motif sets that describe similarities among subgroups of both the related and unrelated proteins. This rich set of motifs give a better description of the similarities and differences between different folds than do previous motif-based methods.",
-
notes = "PMID: 17254344 Undergraduate thesis: Protein Remote
Homology Detection using Motifs made with Genetic
Programming Handstad, Tony. broken
http://urn.ub.uu.se/resolve?urn=urn:nbn:no:ntnu:diva-1030
(2007-03-30) 118 pages.
Binary feature vectors. 2 seconds run time (PC+search chip). GPboost, SCOP, eMOTIF kernel, ROC, classifier combination. 'GPkernel performs significantly better than the other motif-based methods' p5. GPextended. evolves regular expressions. 'In addition to [the 20] amino acid characters, the motifs are also made from the disjunction operator (|) wildcard (.) and Hamming distance {:p>=x} that specifies the minimum number of characters that must match in the pattern.' p13.",
