Thresholding Gini variable importance with a single-trained random forest: an empirical Bayes approach

Robert Dunne*, Roc Reguant, Priya Ramarao-Milne, Piotr Szul, Letitia M. F. Sng, Mischa Lundberg, Natalie A. Twine, Denis C. Bauer*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

Random forests (RFs) are a widely used modelling tool capable of feature selection via a variable importance measure (VIM), however, a threshold is needed to control for false positives. In the absence of a good understanding of the characteristics of VIMs, many current approaches attempt to select features associated to the response by training multiple RFs to generate statistical power via a permutation null, by employing recursive feature elimination, or through a combination of both. However, for high-dimensional datasets these approaches become computationally infeasible. In this paper, we present RFlocalfdr, a statistical approach, built on the empirical Bayes argument of Efron, for thresholding mean decrease in impurity (MDI) importances. It identifies features significantly associated with the response while controlling the false positive rate. Using synthetic data and real-world data in health, we demonstrate that RFlocalfdr has equivalent accuracy to currently published approaches, while being orders of magnitude faster. We show that RFlocalfdr can successfully threshold a dataset of 106 datapoints, establishing its usability for large-scale datasets, like genomics. Furthermore, RFlocalfdr is compatible with any RF implementation that returns a VIM and counts, making it a versatile feature selection tool that reduces false discoveries.

Original languageEnglish
Pages (from-to)4354-4360
Number of pages7
JournalComputational and Structural Biotechnology Journal
Volume21
Early online date1 Sept 2023
DOIs
Publication statusPublished - 2023

Bibliographical note

Copyright the Author(s) 2023. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

  • Empirical Bayes
  • Feature selection
  • Genetic analysis
  • Local FDR
  • Machine learning significance
  • Random forest

Fingerprint

Dive into the research topics of 'Thresholding Gini variable importance with a single-trained random forest: an empirical Bayes approach'. Together they form a unique fingerprint.

Cite this