Limits...
Network and data integration for biomarker signature discovery via network smoothed T-statistics.

Cun Y, Fröhlich H - PLoS ONE (2013)

Bottom Line: We here propose a technique that integrates network information as well as different kinds of experimental data (here exemplified by mRNA and miRNA expression) into one classifier.Moreover, obtained gene lists can be clearly associated to biological knowledge, such as known disease genes and KEGG pathways.We demonstrate that our data integration strategy can improve classification performance compared to using a single data source only.

View Article: PubMed Central - PubMed

Affiliation: Algorithmic Bioinformatics, Bonn-Aachen International Center for IT, Bonn, Germany.

ABSTRACT
Predictive, stable and interpretable gene signatures are generally seen as an important step towards a better personalized medicine. During the last decade various methods have been proposed for that purpose. However, one important obstacle for making gene signatures a standard tool in clinics is the typical low reproducibility of signatures combined with the difficulty to achieve a clear biological interpretation. For that purpose in the last years there has been a growing interest in approaches that try to integrate information from molecular interaction networks. We here propose a technique that integrates network information as well as different kinds of experimental data (here exemplified by mRNA and miRNA expression) into one classifier. This is done by smoothing t-statistics of individual genes or miRNAs over the structure of a combined protein-protein interaction (PPI) and miRNA-target gene network. A permutation test is conducted to select features in a highly consistent manner, and subsequently a Support Vector Machine (SVM) classifier is trained. Compared to several other competing methods our algorithm reveals an overall better prediction performance for early versus late disease relapse and a higher signature stability. Moreover, obtained gene lists can be clearly associated to biological knowledge, such as known disease genes and KEGG pathways. We demonstrate that our data integration strategy can improve classification performance compared to using a single data source only. Our method, called stSVM, is available in R-package netClass on CRAN (http://cran.r-project.org).

Show MeSH

Related in: MedlinePlus

Enrichment of signatures with disease related genes.The y-axis shows - p-values computed via a hypergeometric test (Bonferroni correction for multiple testing). Black horizontal line = 5% significance cutoff.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3760887&req=5

pone-0073074-g003: Enrichment of signatures with disease related genes.The y-axis shows - p-values computed via a hypergeometric test (Bonferroni correction for multiple testing). Black horizontal line = 5% significance cutoff.

Mentions: In order to test the association with existing biological knowledge more systematically we trained each of our tested methods on complete datasets and subsequently tested the resulting signatures (Tables S9, S10, S11, S12 for stSVM, Tables S13 and S14 for stSVM(mi-mRNA) ) for enrichment of disease related genes and KEGG pathways (Figures 3, S2). For testing the association with disease related genes we used the FunDO tool [37], which is based on a hyper-geometric test.


Network and data integration for biomarker signature discovery via network smoothed T-statistics.

Cun Y, Fröhlich H - PLoS ONE (2013)

Enrichment of signatures with disease related genes.The y-axis shows - p-values computed via a hypergeometric test (Bonferroni correction for multiple testing). Black horizontal line = 5% significance cutoff.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3760887&req=5

pone-0073074-g003: Enrichment of signatures with disease related genes.The y-axis shows - p-values computed via a hypergeometric test (Bonferroni correction for multiple testing). Black horizontal line = 5% significance cutoff.
Mentions: In order to test the association with existing biological knowledge more systematically we trained each of our tested methods on complete datasets and subsequently tested the resulting signatures (Tables S9, S10, S11, S12 for stSVM, Tables S13 and S14 for stSVM(mi-mRNA) ) for enrichment of disease related genes and KEGG pathways (Figures 3, S2). For testing the association with disease related genes we used the FunDO tool [37], which is based on a hyper-geometric test.

Bottom Line: We here propose a technique that integrates network information as well as different kinds of experimental data (here exemplified by mRNA and miRNA expression) into one classifier.Moreover, obtained gene lists can be clearly associated to biological knowledge, such as known disease genes and KEGG pathways.We demonstrate that our data integration strategy can improve classification performance compared to using a single data source only.

View Article: PubMed Central - PubMed

Affiliation: Algorithmic Bioinformatics, Bonn-Aachen International Center for IT, Bonn, Germany.

ABSTRACT
Predictive, stable and interpretable gene signatures are generally seen as an important step towards a better personalized medicine. During the last decade various methods have been proposed for that purpose. However, one important obstacle for making gene signatures a standard tool in clinics is the typical low reproducibility of signatures combined with the difficulty to achieve a clear biological interpretation. For that purpose in the last years there has been a growing interest in approaches that try to integrate information from molecular interaction networks. We here propose a technique that integrates network information as well as different kinds of experimental data (here exemplified by mRNA and miRNA expression) into one classifier. This is done by smoothing t-statistics of individual genes or miRNAs over the structure of a combined protein-protein interaction (PPI) and miRNA-target gene network. A permutation test is conducted to select features in a highly consistent manner, and subsequently a Support Vector Machine (SVM) classifier is trained. Compared to several other competing methods our algorithm reveals an overall better prediction performance for early versus late disease relapse and a higher signature stability. Moreover, obtained gene lists can be clearly associated to biological knowledge, such as known disease genes and KEGG pathways. We demonstrate that our data integration strategy can improve classification performance compared to using a single data source only. Our method, called stSVM, is available in R-package netClass on CRAN (http://cran.r-project.org).

Show MeSH
Related in: MedlinePlus