Basic question re: multiple testing
2
0
Entering edit mode
Claire Wilson ▴ 280
@claire-wilson-273
Last seen 10.2 years ago
Dear all, Apologies if this is slightly off-topic/bit basic! With respects to multiple testing, if you apply it does it change the actual order of significance, i.e. are your top 10 most significantly changing genes the same when you perform a t-test with multiple testing and when you apply a t-test without multiple testing? Many thanks in advance Claire -- Claire Wilson Bioinformatics group Paterson Institute for Cancer Research Christies Hospital NHS Trust Wilmslow Road, Withington Manchester M20 4BX tel: +44 (0)161 446 8218 url: http://bioinf.picr.man.ac.uk/ -------------------------------------------------------- This email is confidential and intended solely for the use of th... {{dropped}}
Cancer Cancer • 968 views
ADD COMMENT
0
Entering edit mode
A.J. Rossini ▴ 810
@aj-rossini-209
Last seen 10.2 years ago
"Claire Wilson" <clairewilson@picr.man.ac.uk> writes: > Apologies if this is slightly off-topic/bit basic! With respects to > multiple testing, if you apply it does it change the actual order of > significance, i.e. are your top 10 most significantly changing genes > the same when you perform a t-test with multiple testing and when you > apply a t-test without multiple testing? It depends on the particular adjustment used, and how it rellocates type I error in the hypothesis test. So, yes and no. best, -tony -- A.J. Rossini rossini@u.washington.edu http://software.biostat.washington.edu/ Biostatistics, U Washington and Fred Hutchinson Cancer Research Center FHCRC:Tu: 206-667-7025 (fax=4812)|Voicemail is pretty sketchy/use Email UW : Th: 206-543-1044 (fax=3286)|Change last 4 digits of phone to FAX CONFIDENTIALITY NOTICE: This e-mail message and any attachments ... {{dropped}}
ADD COMMENT
0
Entering edit mode
@adaikalavan-ramasamy-167
Last seen 10.2 years ago
If you use Bonferonni correction, the ranking does not change (strictly speaking when adjusted p-value < 1), because p_adj = min( 1, n*p ). Ie a multiplicative constant n is applied. This is a very conservative estimate as it assumes all the genes are independent of each other. You can learn a lot more about the actual formulae by looking into the codes of p.adjust(). pp <- runif(10000) ppa <- p.adjust( pp, method="fdr") plot(pp, ppa) -----Original Message----- From: Claire Wilson [mailto:ClaireWilson@picr.man.ac.uk] Sent: Friday, May 16, 2003 9:35 PM To: BioC mailing list Subject: [BioC] Basic question re: multiple testing Dear all, Apologies if this is slightly off-topic/bit basic! With respects to multiple testing, if you apply it does it change the actual order of significance, i.e. are your top 10 most significantly changing genes the same when you perform a t-test with multiple testing and when you apply a t-test without multiple testing? Many thanks in advance Claire -- Claire Wilson Bioinformatics group Paterson Institute for Cancer Research Christies Hospital NHS Trust Wilmslow Road, Withington Manchester M20 4BX tel: +44 (0)161 446 8218 url: http://bioinf.picr.man.ac.uk/ -------------------------------------------------------- This email is confidential and intended solely for the use of th... {{dropped}}
ADD COMMENT

Login before adding your answer.

Traffic: 402 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