F Bretz1, J Landgrebe, E Brunner. 1. B and SR, Novartis Pharma AG, Lichtstr. 35, 4002 Basel, Switzerland. frank.bretz@novartis.com
Abstract
OBJECTIVES: Discussion of different error concepts relevant to microarray experiments. Review of some commonly used multiple testing procedures. Comparison of different approaches as applied to gene expression data. METHODS: This article focuses on familywise error rate (FWER) and false discovery rate (FDR) controlling procedures. Methods under investigation include: Bonferroni-type methods and their improvements (including resampling approaches), modified Bonferroni methods, data-driven approaches, as well as the linear step-up method and its modifications. Particular emphasis lies on the description of the assumptions, advantages and limitations for the investigated methods. RESULTS: FWER controlling procedures are often too conservative in high dimensional screening studies. A better balance between the raw P-values and the stringent FWER-adjusted P-values may be required in many situations, as provided by FDR controlling and related procedures. CONCLUSIONS: The questions remain open, which error concept to apply and which multiple testing procedure to use. Although we believe that the FDR or one of its variants will be applied more often in the future, longterm experience with microarray technology is missing and thus the validity of appropriate multiple test procedures cannot yet be assessed for microarray data analysis.
OBJECTIVES: Discussion of different error concepts relevant to microarray experiments. Review of some commonly used multiple testing procedures. Comparison of different approaches as applied to gene expression data. METHODS: This article focuses on familywise error rate (FWER) and false discovery rate (FDR) controlling procedures. Methods under investigation include: Bonferroni-type methods and their improvements (including resampling approaches), modified Bonferroni methods, data-driven approaches, as well as the linear step-up method and its modifications. Particular emphasis lies on the description of the assumptions, advantages and limitations for the investigated methods. RESULTS: FWER controlling procedures are often too conservative in high dimensional screening studies. A better balance between the raw P-values and the stringent FWER-adjusted P-values may be required in many situations, as provided by FDR controlling and related procedures. CONCLUSIONS: The questions remain open, which error concept to apply and which multiple testing procedure to use. Although we believe that the FDR or one of its variants will be applied more often in the future, longterm experience with microarray technology is missing and thus the validity of appropriate multiple test procedures cannot yet be assessed for microarray data analysis.
Authors: Millissia Ben Maamar; Eric Nilsson; Jennifer L M Thorson; Daniel Beck; Michael K Skinner Journal: Reprod Toxicol Date: 2020-09-06 Impact factor: 3.143
Authors: Tatjana Khromov; Ralf Dressel; Iliana Siamishi; Jessica Nolte; Lennart Opitz; Wolfgang Engel; D V Krishna Pantakani Journal: PLoS One Date: 2012-11-07 Impact factor: 3.240
Authors: Soledad Galli; Olaf Jahn; Reiner Hitt; Doerte Hesse; Lennart Opitz; Uwe Plessmann; Henning Urlaub; Juan Jose Poderoso; Elizabeth A Jares-Erijman; Thomas M Jovin Journal: PLoS One Date: 2009-10-22 Impact factor: 3.240
Authors: Tracey E Swingler; Jasmine G Waters; Rosemary K Davidson; Caroline J Pennington; Xose S Puente; Clare Darrah; Adele Cooper; Simon T Donell; Geoffrey R Guile; Wenjia Wang; Ian M Clark Journal: Arthritis Res Ther Date: 2009-06-23 Impact factor: 5.156
Authors: Jyh-Chang Jean; Elizabeth George; Klaus H Kaestner; Lou Ann Scism Brown; Avrum Spira; Martin Joyce-Brady Journal: PLoS One Date: 2013-01-23 Impact factor: 3.240