p.adjust returns one uniform value
1
0
Entering edit mode
Benjamin Otto ▴ 830
@benjamin-otto-1519
Last seen 8.2 years ago
Hi, I'm just sticking on a strange problem. For a set of p-values a correction via p.adjust should be calculated. Strangely the function returns one uniform "adjusted p-value" for all genes although the range of the raw p-values goes from 0.00043 to 0.99. For BH it is something around 0.996 and for holm, bonferroni etc. it is 1. The function produces nice values for the examples and I used to use the function frequently before without ever observing such behaviour. Any suggestions? Benjamin R version 2.5.0 (2007-04-23) i386-pc-mingw32 locale: LC_COLLATE=English_United States.1252;LC_CTYPE=English_United States.1252;LC_MONETARY=English_United States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252 attached base packages: [1] "stats" "graphics" "grDevices" "utils" "datasets" "methods" "base" other attached packages: mouse4302 "1.16.0" ====================================== Benjamin Otto University Hospital Hamburg-Eppendorf Institute For Clinical Chemistry Martinistr. 52 D-20246 Hamburg Tel.: +49 40 42803 1908 Fax.: +49 40 42803 4971 ====================================== -- Pflichtangaben gem?? Gesetz ?ber elektronische Handelsregister und Genossenschaftsregister sowie das Unternehmensregister (EHUG): Universit?tsklinikum Hamburg-Eppendorf K?rperschaft des ?ffentlichen Rechts Gerichtsstand: Hamburg Vorstandsmitglieder: Prof. Dr. J?rg F. Debatin (Vorsitzender) Dr. Alexander Kirstein Ricarda Klein Prof. Dr. Dr. Uwe Koch-Gromus
• 2.7k views
ADD COMMENT
0
Entering edit mode
Jenny Drnevich ★ 2.2k
@jenny-drnevich-382
Last seen 8.2 years ago
Hi Benjamin, There's nothing wrong with p.adjust(), it's your data - there aren't enough genes with low raw p-values. In the BH (fdr) adjustment, you basically take the smallest raw p-value and multiply it by n, the number of genes, to get a temporary adjusted p-value, then you take the second smallest raw p and multiply it by n-1, etc., up the line creating temporary adjusted p-values. When a higher ranking temporary p-value is lower than ones below it, all the p-values below it get replaced with that value as the final adjusted p-value. The largest raw p-value is only multiplied by 1 and in this data set of yours, this is the lowest temporary adjusted p-value for ALL the genes, which is why you're seeing only one uniform "adjusted p-value". Cheers, Jenny At 10:23 AM 9/4/2007, Benjamin Otto wrote: >Hi, > >I'm just sticking on a strange problem. For a set of p-values a correction >via p.adjust should be calculated. Strangely the function returns one >uniform "adjusted p-value" for all genes although the range of the raw >p-values goes from 0.00043 to 0.99. For BH it is something around 0.996 and >for holm, bonferroni etc. it is 1. > >The function produces nice values for the examples and I used to use the >function frequently before without ever observing such behaviour. Any >suggestions? > >Benjamin > > >R version 2.5.0 (2007-04-23) >i386-pc-mingw32 > >locale: >LC_COLLATE=English_United States.1252;LC_CTYPE=English_United >States.1252;LC_MONETARY=English_United >States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252 > >attached base packages: >[1] "stats" "graphics" "grDevices" "utils" "datasets" "methods" >"base" > >other attached packages: >mouse4302 > "1.16.0" > >====================================== >Benjamin Otto >University Hospital Hamburg-Eppendorf >Institute For Clinical Chemistry >Martinistr. 52 >D-20246 Hamburg > >Tel.: +49 40 42803 1908 >Fax.: +49 40 42803 4971 >====================================== > > > >-- >Pflichtangaben gem?? Gesetz ?ber elektronische >Handelsregister und Genossenschaftsregister >sowie das Unternehmensregister (EHUG): > >Universit?tsklinikum Hamburg-Eppendorf >K?rperschaft des ?ffentlichen Rechts >Gerichtsstand: Hamburg > >Vorstandsmitglieder: >Prof. Dr. J?rg F. Debatin (Vorsitzender) >Dr. Alexander Kirstein >Ricarda Klein >Prof. Dr. Dr. Uwe Koch-Gromus > > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor >Search the archives: >http://news.gmane.org/gmane.science.biology.informatics.conductor Jenny Drnevich, Ph.D. Functional Genomics Bioinformatics Specialist W.M. Keck Center for Comparative and Functional Genomics Roy J. Carver Biotechnology Center University of Illinois, Urbana-Champaign 330 ERML 1201 W. Gregory Dr. Urbana, IL 61801 USA ph: 217-244-7355 fax: 217-265-5066 e-mail: drnevich at uiuc.edu
ADD COMMENT
0
Entering edit mode
Hi Jenny, Thanks for the thorough answer. I checked my distributions yesterday and indeed they looked very different to my usual ones, smallest pvalue 0.0004 but about 54000 probesets. :) Best regards Benjamnin -----Urspr?ngliche Nachricht----- Von: Jenny Drnevich [mailto:drnevich at uiuc.edu] Gesendet: Tuesday, September 04, 2007 6:19 PM An: Benjamin Otto; bioconductor at stat.math.ethz.ch Betreff: Re: [BioC] p.adjust returns one uniform value Hi Benjamin, There's nothing wrong with p.adjust(), it's your data - there aren't enough genes with low raw p-values. In the BH (fdr) adjustment, you basically take the smallest raw p-value and multiply it by n, the number of genes, to get a temporary adjusted p-value, then you take the second smallest raw p and multiply it by n-1, etc., up the line creating temporary adjusted p-values. When a higher ranking temporary p-value is lower than ones below it, all the p-values below it get replaced with that value as the final adjusted p-value. The largest raw p-value is only multiplied by 1 and in this data set of yours, this is the lowest temporary adjusted p-value for ALL the genes, which is why you're seeing only one uniform "adjusted p-value". Cheers, Jenny At 10:23 AM 9/4/2007, Benjamin Otto wrote: >Hi, > >I'm just sticking on a strange problem. For a set of p-values a >correction via p.adjust should be calculated. Strangely the function >returns one uniform "adjusted p-value" for all genes although the range >of the raw p-values goes from 0.00043 to 0.99. For BH it is something >around 0.996 and for holm, bonferroni etc. it is 1. > >The function produces nice values for the examples and I used to use >the function frequently before without ever observing such behaviour. >Any suggestions? > >Benjamin > > >R version 2.5.0 (2007-04-23) >i386-pc-mingw32 > >locale: >LC_COLLATE=English_United States.1252;LC_CTYPE=English_United >States.1252;LC_MONETARY=English_United >States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252 > >attached base packages: >[1] "stats" "graphics" "grDevices" "utils" "datasets" "methods" >"base" > >other attached packages: >mouse4302 > "1.16.0" > >====================================== >Benjamin Otto >University Hospital Hamburg-Eppendorf >Institute For Clinical Chemistry >Martinistr. 52 >D-20246 Hamburg > >Tel.: +49 40 42803 1908 >Fax.: +49 40 42803 4971 >====================================== > > > >-- >Pflichtangaben gem?? Gesetz ?ber elektronische Handelsregister und >Genossenschaftsregister sowie das Unternehmensregister (EHUG): > >Universit?tsklinikum Hamburg-Eppendorf >K?rperschaft des ?ffentlichen Rechts >Gerichtsstand: Hamburg > >Vorstandsmitglieder: >Prof. Dr. J?rg F. Debatin (Vorsitzender) Dr. Alexander Kirstein Ricarda >Klein Prof. Dr. Dr. Uwe Koch-Gromus > > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor >Search the archives: >http://news.gmane.org/gmane.science.biology.informatics.conductor Jenny Drnevich, Ph.D. Functional Genomics Bioinformatics Specialist W.M. Keck Center for Comparative and Functional Genomics Roy J. Carver Biotechnology Center University of Illinois, Urbana-Champaign 330 ERML 1201 W. Gregory Dr. Urbana, IL 61801 USA ph: 217-244-7355 fax: 217-265-5066 e-mail: drnevich at uiuc.edu -- Pflichtangaben gem?? Gesetz ?ber elektronische Handelsregister und Genossenschaftsregister sowie das Unternehmensregister (EHUG): Universit?tsklinikum Hamburg-Eppendorf K?rperschaft des ?ffentlichen Rechts Gerichtsstand: Hamburg Vorstandsmitglieder: Prof. Dr. J?rg F. Debatin (Vorsitzender) Dr. Alexander Kirstein Ricarda Klein Prof. Dr. Dr. Uwe Koch-Gromus
ADD REPLY

Login before adding your answer.

Traffic: 614 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6