Bayesian automatic relevance determination algorithms for classifying gene expression data

• Published in: Bioinformatics (Oxford, England) Vol. 18; no. 10; pp. 1332 - 1339
• Main Authors: Li, Yi, Campbell, Colin, Tipping, Michael
• Format: Journal Article
• Language: English
• Published: Oxford Oxford University Press 01.10.2002
• Subjects: Algorithms; Bacteriology; Bayes Theorem; Biological and medical sciences; Cluster Analysis; Colonic Neoplasms - classification; Colonic Neoplasms - genetics; Databases, Nucleic Acid; Female; Fundamental and applied biological sciences. Psychology; Gene Expression Profiling - methods; Gene Expression Regulation, Neoplastic; General aspects; Genetic Markers - genetics; Genetics; Humans; Leukemia, Myeloid, Acute - classification; Leukemia, Myeloid, Acute - genetics; Mathematics in biology. Statistical analysis. Models. Metrology. Data processing in biology (general aspects); Microbiology; Models, Genetic; Neoplasms - classification; Neoplasms - genetics; Oligonucleotide Array Sequence Analysis - methods; Ovarian Neoplasms - classification; Ovarian Neoplasms - genetics; Pathogenicity, virulence, toxins, bacteriocins, pyrogens, host-bacteria relations, miscellaneous strains; Pattern Recognition, Automated; Precursor Cell Lymphoblastic Leukemia-Lymphoma - classification; Precursor Cell Lymphoblastic Leukemia-Lymphoma - genetics; Quality Control; Reproducibility of Results; Sensitivity and Specificity; Sequence Analysis, DNA - methods; Statistical analysis; Complementary DNA; Genomics; Classification; DNA chip; Bayes network; Malignant tumor; Gene expression; Bioinformatics; Algorithm
• ISSN: 1367-4803; 1367-4811
• DOI: 10.1093/bioinformatics/18.10.1332

Motivation: We investigate two new Bayesian classification algorithms incorporating feature selection. These algorithms are applied to the classification of gene expression data derived from cDNA microarrays. Results: We demonstrate the effectiveness of the algorithms on three gene expression datasets for cancer, showing they compare well with alternative kernel-based techniques. By automatically incorporating feature selection, accurate classifiers can be constructed utilizing very few features and with minimal hand-tuning. We argue that the feature selection is meaningful and some of the highlighted genes appear to be medically important. Contact: C.Campbell@bris.ac.uk * To whom correspondence should be addressed.  Present address: Information and Mathematical Sciences, Genome Institute of Singapore, 1 Science Park Road, The Capricorn #05-01, Singapore 117528, Republic of Singapore