English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Motivation: Biomarker discovery from high-dimensional data is a crucialproblem with enormous applications in biology and medicine. It is alsoextremely challenging from a statistical viewpoint, but surprisingly fewstudies have investigated the relative strengths and weaknesses of the plethoraof existing feature selection methods. Methods: We compare 32 feature selectionmethods on 4 public gene expression datasets for breast cancer prognosis, interms of predictive performance, stability and functional interpretability ofthe signatures they produce. Results: We observe that the feature selectionmethod has a significant influence on the accuracy, stability andinterpretability of signatures. Simple filter methods generally outperform morecomplex embedded or wrapper methods, and ensemble feature selection hasgenerally no positive effect. Overall a simple Student's t-test seems toprovide the best results. Availability: Code and data are publicly available athttp://cbio.ensmp.fr/~ahaury/.

The influence of feature selection methods on accuracy, stability and interpretability of molecular signatures