RT Conference Proceedings
T1 Machine learning models to search relevant genetic signatures in clinical context
A1 Urda, Daniel
A1 Luque-Baena, Rafael Marcos
A1 Franco, Leonardo
A1 Sánchez-Maroño, Noelia
A1 Jerez-Aragonés, José Manuel
K1 Bioinformática
K1 Biología computacional
AB Clinicians are interested in the estimation of robustand relevant genetic signatures from gene sequencing data.Many machine learning approaches have been proposed tryingto address well-known issues of this complex task (feature orgene selection, classification or model selection, and predictionassessment). Addressing this problem often requires a deepknowledge of these methods and some of them demand highcomputational resources that may not be affordable. In thispaper, an exhaustive study that includes different types of featureselection methods and classifiers is presented, providing clinicians an useful insight of the most suitable methods for this purpose. Predictions assessment is performed using a bootstrap crossvalidation strategy as an honest validation scheme. The results of this study for six benchmark datasets show that filter or embedded methods are preferred, in general, to wrapper methodsaccording to their better statistical significant results, in termsof accuracy, and lower demand for computational resources.
YR 2017
FD 2017-06-26
LK http://hdl.handle.net/10630/14000
UL http://hdl.handle.net/10630/14000
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 13 abr 2026