Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations

Abstract The false discovery rate (FDR) controlling method by Benjamini and Hochberg (BH) is a popular choice in the omics fields. Here, we demonstrate that in datasets with a large degree of dependencies between features, FDR correction methods like BH can sometimes counter-intuitively report very...

Full description

Saved in:
Bibliographic Details
Main Authors: Chakravarthi Kanduri, Maria Mamica, Emilie Willoch Olstad, Manuela Zucknick, Jingyi Jessica Li, Geir Kjetil Sandve
Format: Article
Language:English
Published: BMC 2025-08-01
Series:Genome Biology
Subjects:
Online Access:https://doi.org/10.1186/s13059-025-03734-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849330490031472640
author Chakravarthi Kanduri
Maria Mamica
Emilie Willoch Olstad
Manuela Zucknick
Jingyi Jessica Li
Geir Kjetil Sandve
author_facet Chakravarthi Kanduri
Maria Mamica
Emilie Willoch Olstad
Manuela Zucknick
Jingyi Jessica Li
Geir Kjetil Sandve
author_sort Chakravarthi Kanduri
collection DOAJ
description Abstract The false discovery rate (FDR) controlling method by Benjamini and Hochberg (BH) is a popular choice in the omics fields. Here, we demonstrate that in datasets with a large degree of dependencies between features, FDR correction methods like BH can sometimes counter-intuitively report very high numbers of false positives, potentially misleading researchers. We call the attention of researchers to use suited multiple testing strategies and approaches like synthetic null data (negative control) to identify and minimize caveats related to false discoveries, as in the cases where false findings do occur, they may be numerous.
format Article
id doaj-art-b2b453fd8bd04b4990dc4e7e852a7e8a
institution Kabale University
issn 1474-760X
language English
publishDate 2025-08-01
publisher BMC
record_format Article
series Genome Biology
spelling doaj-art-b2b453fd8bd04b4990dc4e7e852a7e8a2025-08-20T03:46:54ZengBMCGenome Biology1474-760X2025-08-0126111710.1186/s13059-025-03734-zBeware of counter-intuitive levels of false discoveries in datasets with strong intra-correlationsChakravarthi Kanduri0Maria Mamica1Emilie Willoch Olstad2Manuela Zucknick3Jingyi Jessica Li4Geir Kjetil Sandve5Scientific Computing and Machine Learning Section, Department of Informatics, University of OsloScientific Computing and Machine Learning Section, Department of Informatics, University of OsloUiORealArt Convergence Environment, University of OsloDepartment of Biostatistics, Faculty of Medicine, University of OsloDepartment of Statistics and Data Science, University of CaliforniaScientific Computing and Machine Learning Section, Department of Informatics, University of OsloAbstract The false discovery rate (FDR) controlling method by Benjamini and Hochberg (BH) is a popular choice in the omics fields. Here, we demonstrate that in datasets with a large degree of dependencies between features, FDR correction methods like BH can sometimes counter-intuitively report very high numbers of false positives, potentially misleading researchers. We call the attention of researchers to use suited multiple testing strategies and approaches like synthetic null data (negative control) to identify and minimize caveats related to false discoveries, as in the cases where false findings do occur, they may be numerous.https://doi.org/10.1186/s13059-025-03734-zFalse discovery rate (FDR)Benjamini-Hochberg (BH)Multiple testing adjustmentHigh-dimensional omics data
spellingShingle Chakravarthi Kanduri
Maria Mamica
Emilie Willoch Olstad
Manuela Zucknick
Jingyi Jessica Li
Geir Kjetil Sandve
Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
Genome Biology
False discovery rate (FDR)
Benjamini-Hochberg (BH)
Multiple testing adjustment
High-dimensional omics data
title Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
title_full Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
title_fullStr Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
title_full_unstemmed Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
title_short Beware of counter-intuitive levels of false discoveries in datasets with strong intra-correlations
title_sort beware of counter intuitive levels of false discoveries in datasets with strong intra correlations
topic False discovery rate (FDR)
Benjamini-Hochberg (BH)
Multiple testing adjustment
High-dimensional omics data
url https://doi.org/10.1186/s13059-025-03734-z
work_keys_str_mv AT chakravarthikanduri bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations
AT mariamamica bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations
AT emiliewillocholstad bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations
AT manuelazucknick bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations
AT jingyijessicali bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations
AT geirkjetilsandve bewareofcounterintuitivelevelsoffalsediscoveriesindatasetswithstrongintracorrelations