EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Statistical knockoffs improve biomarker discovery from transcriptomic data

Julie Cartier iD, Johanna Lagoas, Youmna Ayadi, Adeline Fermanian, Chloé-Agathe Azencott iD, Florian Massip iD

Briefings in Bioinformatics · 2026

DOI: 10.1093/bib/bbag148 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Julie Cartier und 5 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Computational Biology, Mines Paris, PSL University, 60 bd Saint-Michel, 75272 Paris, France. Hauptthemen laut MEDLINE: Transcriptome, Gene Expression Profiling, Biomarkers. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: French Agence Nationale de la Recherche.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Advances in sequencing technologies have enabled the generation of large amounts of data, offering new possibilities to identify relationships between biological units (e.g. genes) and phenotypic traits (e.g. disease outcomes). Yet, identifying these associations using variable selection methods remains challenging due to the high dimension ($p \gg n$) and the correlation structure of the data. To address these challenges, we study the applicability of the knockoff (KO) procedure. Introduced by Barber and Candès in 2015, the KO variable selection procedure has shown promising results on real biological data, such as Genome-Wide Association Studies. This method seeks to identify the truly important predictors by overcoming the correlation structure between variables while controlling the false discovery rate. Here, we study the applicability of the KO procedure on transcriptomic data in a classification setting. We conduct an extensive simulation study using real transcriptomic data to evaluate the performance of the KO framework in the context of high-dimensional classification. We find that the KO framework outperforms widely used variable selection models, and that using KO aggregation to mitigate the effect of KO stochasticity improves stability while maintaining the same power. Finally, applied to three real transcriptomic datasets, the KO framework made very few discoveries, highlighting its conservative nature and suggesting that other methods may substantially overestimate the number of relevant features.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

TranscriptomeGene Expression ProfilingBiomarkersGenome-Wide Association StudyHumansComputer SimulationComputational BiologyAlgorithms

Stichwörter der Autor:innen

knockoffstranscriptomicvariable selection

Substanzen

Biomarkers

Autor:innen und Institutionen

  • Centre for Computational Biology, Mines Paris, PSL University, 60 bd Saint-Michel, 75272 Paris, France.
  • Institut Curie, PSL University, 11 rue Pierre et Marie Curie, 75005 Paris, France.
  • U1331, INSERM, 11 rue Pierre et Marie Curie, 75005 Paris, France.
  • LOPF, LOPF Califrais'Machine Learning Lab, 4 Quai du Val de Loire, 94550 Chevilly Larue, France.

ORCID-Profile: Julie Cartier, Chloé-Agathe Azencott, Florian Massip

Förderung

  • French Agence Nationale de la Recherche

Zitieren

APA

Cartier, J., Lagoas, J., Ayadi, Y., Fermanian, A., Azencott, C. A., & Massip, F. (2026). Statistical knockoffs improve biomarker discovery from transcriptomic data. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag148

Vancouver

Cartier J, Lagoas J, Ayadi Y, Fermanian A, Azencott CA, Massip F. Statistical knockoffs improve biomarker discovery from transcriptomic data. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag148

BibTeX
@article{Cartier2026,
  author = {Cartier, Julie and Lagoas, Johanna and Ayadi, Youmna and Fermanian, Adeline and Azencott, Chloé-Agathe and Massip, Florian},
  title = {Statistical knockoffs improve biomarker discovery from transcriptomic data},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  doi = {10.1093/bib/bbag148}
}
Kostenlos starten →