A direct approach to estimating false discovery rates conditional on covariates
Modern scientific studies from many diverse areas of research abound with multiple hypothesis testing concerns. The false discovery rate (FDR) is one of the most commonly used approaches for measuring and controlling error rates when performing multiple tests. Adaptive FDRs rely on an estimate of th...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2018-12-01
|
| Series: | PeerJ |
| Subjects: | |
| Online Access: | https://peerj.com/articles/6035.pdf |
| _version_ | 1797418611210452992 |
|---|---|
| author | Simina M. Boca Jeffrey T. Leek |
| author_facet | Simina M. Boca Jeffrey T. Leek |
| author_sort | Simina M. Boca |
| collection | DOAJ |
| description | Modern scientific studies from many diverse areas of research abound with multiple hypothesis testing concerns. The false discovery rate (FDR) is one of the most commonly used approaches for measuring and controlling error rates when performing multiple tests. Adaptive FDRs rely on an estimate of the proportion of null hypotheses among all the hypotheses being tested. This proportion is typically estimated once for each collection of hypotheses. Here, we propose a regression framework to estimate the proportion of null hypotheses conditional on observed covariates. This may then be used as a multiplication factor with the Benjamini–Hochberg adjusted p-values, leading to a plug-in FDR estimator. We apply our method to a genome-wise association meta-analysis for body mass index. In our framework, we are able to use the sample sizes for the individual genomic loci and the minor allele frequencies as covariates. We further evaluate our approach via a number of simulation scenarios. We provide an implementation of this novel method for estimating the proportion of null hypotheses in a regression framework as part of the Bioconductor package swfdr. |
| first_indexed | 2024-03-09T06:36:24Z |
| format | Article |
| id | doaj.art-a43f741641dc4fac83bcf452fe7f11a0 |
| institution | Directory Open Access Journal |
| issn | 2167-8359 |
| language | English |
| last_indexed | 2024-03-09T06:36:24Z |
| publishDate | 2018-12-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ |
| spelling | doaj.art-a43f741641dc4fac83bcf452fe7f11a02023-12-03T10:58:18ZengPeerJ Inc.PeerJ2167-83592018-12-016e603510.7717/peerj.6035A direct approach to estimating false discovery rates conditional on covariatesSimina M. Boca0Jeffrey T. Leek1Innovation Center for Biomedical Informatics, Georgetown University Medical Center, Washington, D.C., USADepartment of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USAModern scientific studies from many diverse areas of research abound with multiple hypothesis testing concerns. The false discovery rate (FDR) is one of the most commonly used approaches for measuring and controlling error rates when performing multiple tests. Adaptive FDRs rely on an estimate of the proportion of null hypotheses among all the hypotheses being tested. This proportion is typically estimated once for each collection of hypotheses. Here, we propose a regression framework to estimate the proportion of null hypotheses conditional on observed covariates. This may then be used as a multiplication factor with the Benjamini–Hochberg adjusted p-values, leading to a plug-in FDR estimator. We apply our method to a genome-wise association meta-analysis for body mass index. In our framework, we are able to use the sample sizes for the individual genomic loci and the minor allele frequencies as covariates. We further evaluate our approach via a number of simulation scenarios. We provide an implementation of this novel method for estimating the proportion of null hypotheses in a regression framework as part of the Bioconductor package swfdr.https://peerj.com/articles/6035.pdfFalse discovery ratesFDR regressionAdaptive FDR |
| spellingShingle | Simina M. Boca Jeffrey T. Leek A direct approach to estimating false discovery rates conditional on covariates PeerJ False discovery rates FDR regression Adaptive FDR |
| title | A direct approach to estimating false discovery rates conditional on covariates |
| title_full | A direct approach to estimating false discovery rates conditional on covariates |
| title_fullStr | A direct approach to estimating false discovery rates conditional on covariates |
| title_full_unstemmed | A direct approach to estimating false discovery rates conditional on covariates |
| title_short | A direct approach to estimating false discovery rates conditional on covariates |
| title_sort | direct approach to estimating false discovery rates conditional on covariates |
| topic | False discovery rates FDR regression Adaptive FDR |
| url | https://peerj.com/articles/6035.pdf |
| work_keys_str_mv | AT siminamboca adirectapproachtoestimatingfalsediscoveryratesconditionaloncovariates AT jeffreytleek adirectapproachtoestimatingfalsediscoveryratesconditionaloncovariates AT siminamboca directapproachtoestimatingfalsediscoveryratesconditionaloncovariates AT jeffreytleek directapproachtoestimatingfalsediscoveryratesconditionaloncovariates |