Novel gene sets improve set-level classification of prokaryotic gene expression data

BackgroundSet-level classification of gene expression data has received significant attention recently. In this setting, high-dimensional vectors of features corresponding to genes are converted into lower-dimensional vectors of features corresponding to biologically interpretable gene sets. The dimensionality reduction brings the promise of a decreased risk of overfitting, potentially resulting in improved accuracy of the learned classifiers. However, recent empirical research has not confirmed this expectation. Here we hypothesize that the reported unfavorable classification results in the set-level framework were due to the adoption of unsuitable gene sets defined typically on the basis of the Gene ontology and the KEGG database of metabolic networks. We explore an alternative approach to defining gene sets, based on regulatory interactions, which we expect to collect genes with more correlated expression. We hypothesize that such more correlated gene sets will enable to learn more accurate classifiers.MethodsWe define two families of gene sets using information on regulatory interactions, and evaluate them on phenotype-classification tasks using public prokaryotic gene expression data sets. From each of the two gene-set families, we first select the best-performing subtype. The two selected subtypes are then evaluated on independent (testing) data sets against state-of-the-art gene sets and against the conventional gene-level approach.ResultsThe novel gene sets are indeed more correlated than the conventional ones, and lead to significantly more accurate classifiers. The novel gene sets are indeed more correlated than the conventional ones, and lead to significantly more accurate classifiers.ConclusionNovel gene sets defined on the basis of regulatory interactions improve set-level classification of gene expression data. The experimental scripts and other material needed to reproduce the experiments are available at http://ida.felk.cvut.cz/novelgenesets.tar.gz.

[1]  J. Collado-Vides,et al.  The repertoire of DNA-binding transcriptional regulators in Escherichia coli K-12. , 2000, Nucleic acids research.

[2]  Blaz Zupan,et al.  On utility of gene set signatures in gene expression-based cancer class prediction , 2009, MLSB.

[3]  Filip Zelezný,et al.  Comparative evaluation of set-level techniques in predictive classification of gene expression samples , 2012, BMC Bioinformatics.

[4]  Jiří Kléma,et al.  Network-constrained forest for regularized classification of omics data. , 2015, Methods.

[5]  M. Ashburner,et al.  Gene Ontology: tool for the unification of biology , 2000, Nature Genetics.

[6]  Alex E. Lash,et al.  Gene Expression Omnibus: NCBI gene expression and hybridization array data repository , 2002, Nucleic Acids Res..

[7]  R. Sram,et al.  Genotoxicity but not the AhR-mediated activity of PAHs is inhibited by other components of complex mixtures of ambient air pollutants. , 2014, Toxicology letters.

[8]  Matthew DeJongh,et al.  Evaluating the consistency of gene sets used in the analysis of bacterial gene expression data , 2011, BMC Bioinformatics.

[9]  Wei Pan,et al.  Operon information improves gene expression estimation for cDNA microarrays , 2006, BMC Genomics.

[10]  Z. Zemanová,et al.  Genome‐wide miRNA profiling in myelodysplastic syndrome with del(5q) treated with lenalidomide , 2015, European journal of haematology.

[11]  Bruno Crémilleux,et al.  Mining Plausible Patterns from Genomic Data , 2006, 19th IEEE Symposium on Computer-Based Medical Systems (CBMS'06).

[12]  Lodewyk F. A. Wessels,et al.  Current composite-feature classification methods do not outperform simple single-genes classifiers in breast cancer prognosis , 2013, Front. Genet..

[13]  Céline Robardet,et al.  SQUAT: A web tool to mine human, murine and avian SAGE data , 2008, BMC Bioinformatics.

[14]  Susumu Goto,et al.  The KEGG resource for deciphering the genome , 2004, Nucleic Acids Res..

[15]  Sean R. Davis,et al.  NCBI GEO: archive for functional genomics data sets—update , 2012, Nucleic Acids Res..

[16]  Araceli M. Huerta,et al.  Regulatory network of Escherichia coli: consistency between literature knowledge and microarray profiles. , 2003, Genome research.

[17]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[18]  Sushmita Mitra,et al.  Feature Selection and Clustering of Gene Expression Profiles Using Biological Knowledge , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[19]  S. Tenenbaum,et al.  Eukaryotic mRNPs may represent posttranscriptional operons. , 2002, Molecular cell.

[20]  Brad T. Sherman,et al.  Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists , 2008, Nucleic acids research.

[21]  Julio Collado-Vides,et al.  RegulonDB v8.0: omics data sets, evolutionary conservation, regulatory phrases, cross-validated gold standards and more , 2012, Nucleic Acids Res..

[22]  Janez Demsar,et al.  Statistical Comparisons of Classifiers over Multiple Data Sets , 2006, J. Mach. Learn. Res..

[23]  Justin Zobel,et al.  Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context , 2010, BMC Bioinformatics.

[24]  W. Maas,et al.  STUDIES ON THE MECHANISM OF REPRESSION OF ARGININE BIOSYNTHESIS IN ESCHERICHIA COLI. II. DOMINANCE OF REPRESSIBILITY IN DIPLOIDS. , 1964, Journal of molecular biology.

[25]  Jirí Kléma,et al.  Empirical Evidence of the Applicability of Functional Clustering through Gene Expression Classification , 2012, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[26]  Lodewyk F. A. Wessels,et al.  A Critical Evaluation of Network and Pathway-Based Classifiers for Outcome Prediction in Breast Cancer , 2011, PloS one.