On utility of gene set signatures in gene expression-based cancer class prediction

Minca Mramor, Marko Toplak, Gregor Leban, Tomaž Curk, Janez Demšar, Blaž Zupan
Proceedings of the third International Workshop on Machine Learning in Systems Biology, PMLR 8:55-64, 2009.

Abstract

Machine learning methods that can use additional knowledge in their inference process are central to the development of integrative bioinformatics. Inclusion of background knowledge improves robustness, predictive accuracy and interpretability. Recently, a set of such techniques has been proposed that use information on gene sets for supervised data mining of class-labeled microarray data sets. We here present a new gene set-based supervised learning approach named SetSig and systematically investigate the predictive accuracy of this and other gene set approaches compared to the standard inference model where only gene expression information is used. Our results indicate that SetSig outperforms other gene set approaches, but contrary to earlier reports, transformation of gene expression data to the space of gene set signatures does not result in increased accuracy of predictive models when compared to those trained directly from original (not transformed) data.

Cite this Paper


BibTeX
@InProceedings{pmlr-v8-mramor10a, title = {On utility of gene set signatures in gene expression-based cancer class prediction}, author = {Mramor, Minca and Toplak, Marko and Leban, Gregor and Curk, Tomaž and Demšar, Janez and Zupan, Blaž}, booktitle = {Proceedings of the third International Workshop on Machine Learning in Systems Biology}, pages = {55--64}, year = {2009}, editor = {Džeroski, Sašo and Guerts, Pierre and Rousu, Juho}, volume = {8}, series = {Proceedings of Machine Learning Research}, address = {Ljubljana, Slovenia}, month = {05--06 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v8/mramor10a/mramor10a.pdf}, url = {https://proceedings.mlr.press/v8/mramor10a.html}, abstract = {Machine learning methods that can use additional knowledge in their inference process are central to the development of integrative bioinformatics. Inclusion of background knowledge improves robustness, predictive accuracy and interpretability. Recently, a set of such techniques has been proposed that use information on gene sets for supervised data mining of class-labeled microarray data sets. We here present a new gene set-based supervised learning approach named SetSig and systematically investigate the predictive accuracy of this and other gene set approaches compared to the standard inference model where only gene expression information is used. Our results indicate that SetSig outperforms other gene set approaches, but contrary to earlier reports, transformation of gene expression data to the space of gene set signatures does not result in increased accuracy of predictive models when compared to those trained directly from original (not transformed) data.} }
Endnote
%0 Conference Paper %T On utility of gene set signatures in gene expression-based cancer class prediction %A Minca Mramor %A Marko Toplak %A Gregor Leban %A Tomaž Curk %A Janez Demšar %A Blaž Zupan %B Proceedings of the third International Workshop on Machine Learning in Systems Biology %C Proceedings of Machine Learning Research %D 2009 %E Sašo Džeroski %E Pierre Guerts %E Juho Rousu %F pmlr-v8-mramor10a %I PMLR %P 55--64 %U https://proceedings.mlr.press/v8/mramor10a.html %V 8 %X Machine learning methods that can use additional knowledge in their inference process are central to the development of integrative bioinformatics. Inclusion of background knowledge improves robustness, predictive accuracy and interpretability. Recently, a set of such techniques has been proposed that use information on gene sets for supervised data mining of class-labeled microarray data sets. We here present a new gene set-based supervised learning approach named SetSig and systematically investigate the predictive accuracy of this and other gene set approaches compared to the standard inference model where only gene expression information is used. Our results indicate that SetSig outperforms other gene set approaches, but contrary to earlier reports, transformation of gene expression data to the space of gene set signatures does not result in increased accuracy of predictive models when compared to those trained directly from original (not transformed) data.
RIS
TY - CPAPER TI - On utility of gene set signatures in gene expression-based cancer class prediction AU - Minca Mramor AU - Marko Toplak AU - Gregor Leban AU - Tomaž Curk AU - Janez Demšar AU - Blaž Zupan BT - Proceedings of the third International Workshop on Machine Learning in Systems Biology DA - 2009/03/02 ED - Sašo Džeroski ED - Pierre Guerts ED - Juho Rousu ID - pmlr-v8-mramor10a PB - PMLR DP - Proceedings of Machine Learning Research VL - 8 SP - 55 EP - 64 L1 - http://proceedings.mlr.press/v8/mramor10a/mramor10a.pdf UR - https://proceedings.mlr.press/v8/mramor10a.html AB - Machine learning methods that can use additional knowledge in their inference process are central to the development of integrative bioinformatics. Inclusion of background knowledge improves robustness, predictive accuracy and interpretability. Recently, a set of such techniques has been proposed that use information on gene sets for supervised data mining of class-labeled microarray data sets. We here present a new gene set-based supervised learning approach named SetSig and systematically investigate the predictive accuracy of this and other gene set approaches compared to the standard inference model where only gene expression information is used. Our results indicate that SetSig outperforms other gene set approaches, but contrary to earlier reports, transformation of gene expression data to the space of gene set signatures does not result in increased accuracy of predictive models when compared to those trained directly from original (not transformed) data. ER -
APA
Mramor, M., Toplak, M., Leban, G., Curk, T., Demšar, J. & Zupan, B.. (2009). On utility of gene set signatures in gene expression-based cancer class prediction. Proceedings of the third International Workshop on Machine Learning in Systems Biology, in Proceedings of Machine Learning Research 8:55-64 Available from https://proceedings.mlr.press/v8/mramor10a.html.

Related Material