Filtering genes to improve sensitivity in oligonucleotide microarray data analysis. Export

@article{citeulike:1579883, abstract = {Many recent microarrays hold an enormous number of probe sets, thus raising many practical and theoretical problems in controlling the false discovery rate (FDR). Biologically, it is likely that most probe sets are associated with un-expressed genes, so the measured values are simply noise due to non-specific binding; also many probe sets are associated with non-differentially-expressed (non-DE) genes. In an analysis to find DE genes, these probe sets contribute to the false discoveries, so it is desirable to filter out these probe sets prior to analysis. In the methodology proposed here, we first fit a robust linear model for probe-level Affymetrix data that accounts for probe and array effects. We then develop a novel procedure called FLUSH (Filtering Likely Uninformative Sets of Hybridizations), which excludes probe sets that have statistically small array-effects or large residual variance. This filtering procedure was evaluated on a publicly available data set from a controlled spiked-in experiment, as well as on a real experimental data set of a mouse model for retinal degeneration. In both cases, FLUSH filtering improves the sensitivity in the detection of DE genes compared to analyses using unfiltered, presence-filtered, intensity-filtered and variance-filtered data. A freely-available package called FLUSH implements the procedures and graphical displays described in the article.}, address = {Department of Medical Epidemiology and Biostatistics - Karolinska Institute, Stockholm, Sweden, Section of Medical Statistics and Biometry, Department of Biomedical Sciences and Biotechnology - University of Brescia, Italy, Laboratoire de Bioinformatique et G\'{e}nomique Int\'{e}gratives, Institut de G\'{e}n\'{e}tique et de Biologie Mol\'{e}culaire et Cellulaire (IGBMC), Illkirch Strasbourg, France and Laboratoire de Physiopathologie Cellulaire et Mol\'{e}culaire de la Retine - Facult\'{e} de M\'{e}decine, Universit\'{e} Pierre et Marie Curie, Paris, France.}, author = {Calza, Stefano and Raffelsberger, Wolfgang and Ploner, Alexander and Sahel, Jose and Leveillard, Thierry and Pawitan, Yudi}, citeulike-article-id = {1579883}, citeulike-linkout-0 = {http://dx.doi.org/10.1093/nar/gkm537}, citeulike-linkout-1 = {http://nar.oxfordjournals.org/cgi/content/abstract/35/16/e102}, citeulike-linkout-2 = {http://view.ncbi.nlm.nih.gov/pubmed/17702762}, citeulike-linkout-3 = {http://www.hubmed.org/display.cgi?uids=17702762}, day = {15}, doi = {10.1093/nar/gkm537}, issn = {1362-4962}, journal = {Nucleic Acids Res}, keywords = {absent-present-calls, bioconductor, flush, gen-filtering, microarray-data-analysis, variance, variance-filtering}, month = {August}, posted-at = {2008-09-04 14:33:32}, priority = {2}, title = {Filtering genes to improve sensitivity in oligonucleotide microarray data analysis.}, url = {http://dx.doi.org/10.1093/nar/gkm537}, year = {2007} }

TY - JOUR ID - citeulike:1579883 L3 - citeulike-article-id:1579883 N2 - Many recent microarrays hold an enormous number of probe sets, thus raising many practical and theoretical problems in controlling the false discovery rate (FDR). Biologically, it is likely that most probe sets are associated with un-expressed genes, so the measured values are simply noise due to non-specific binding; also many probe sets are associated with non-differentially-expressed (non-DE) genes. In an analysis to find DE genes, these probe sets contribute to the false discoveries, so it is desirable to filter out these probe sets prior to analysis. In the methodology proposed here, we first fit a robust linear model for probe-level Affymetrix data that accounts for probe and array effects. We then develop a novel procedure called FLUSH (Filtering Likely Uninformative Sets of Hybridizations), which excludes probe sets that have statistically small array-effects or large residual variance. This filtering procedure was evaluated on a publicly available data set from a controlled spiked-in experiment, as well as on a real experimental data set of a mouse model for retinal degeneration. In both cases, FLUSH filtering improves the sensitivity in the detection of DE genes compared to analyses using unfiltered, presence-filtered, intensity-filtered and variance-filtered data. A freely-available package called FLUSH implements the procedures and graphical displays described in the article. JF - Nucleic Acids Res SN - 1362-4962 AD - Department of Medical Epidemiology and Biostatistics - Karolinska Institute, Stockholm, Sweden, Section of Medical Statistics and Biometry, Department of Biomedical Sciences and Biotechnology - University of Brescia, Italy, Laboratoire de Bioinformatique et Génomique Intégratives, Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC), Illkirch Strasbourg, France and Laboratoire de Physiopathologie Cellulaire et Moléculaire de la Retine - Faculté de Médecine, Université Pierre et Marie Curie, Paris, France. TI - Filtering genes to improve sensitivity in oligonucleotide microarray data analysis. KW - absent-present-calls KW - bioconductor KW - flush KW - gen-filtering KW - microarray-data-analysis KW - variance KW - variance-filtering AU - Calza, Stefano AU - Raffelsberger, Wolfgang AU - Ploner, Alexander AU - Sahel, Jose AU - Leveillard, Thierry AU - Pawitan, Yudi PY - 2007/08/15/ UR - http://dx.doi.org/10.1093/nar/gkm537 ER -