Hello, all I am using Limma for my array analysis. Based on the user's guide, I canconvert B value to probability. however, in my results, I have adj.P.Val of 0.007 with B negative. If I use the conversion, the probability will be 0.25. But when you look at the p value, it passed the significant 0.05 level. How can I explain that? Did I misunderstand some points? Thanks Sabrina

Dear Sabrina I suspect that the question below is too vague to encourage answers. Have a look at the posting guide, and try to give a reproducible example and sessionInfo(). Best wishes Wolfgang Sabrina AT Case scripsit 17/12/10 19:59: > Hello, all > I am using Limma for my array analysis. Based on the user's guide, I > canconvert B value to probability. > however, in my results, I have adj.P.Val of 0.007 with B negative. If > I use the conversion, the probability will be 0.25. But when you look > at the p value, it passed the significant 0.05 level. How can I > explain that? Did I misunderstand some points? Thanks > > Sabrina > -- Wolfgang Huber EMBL http://www.embl.de/research/units/genome_biology/huber

Sabrina AT Case <hxs61 at="" ...=""> writes: > > Hello, all > I am using Limma for my array analysis. Based on the user's guide, I > canconvert B value to probability. > however, in my results, I have adj.P.Val of 0.007 with B negative. If > I use the conversion, the probability will be 0.25. But when you look > at the p value, it passed the significant 0.05 level. How can I > explain that? Did I misunderstand some points? Thanks > > Sabrina > > _______________________________________________ > Bioconductor mailing list > Bioconductor at ... > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor > > hi, It looks to me like you are confusing the p (or adj p) value with the B-stat probability. The (adj) p-value reported in the topTable function is associated with the mod-t statistic and is not related to the B-stat probability. The B-stat probability is obtained by converting the log odds ratio to obtain the probability p. Does this clear up your question? Best regards, Darlene -- Darlene Goldstein ?cole Polytechnique F?d?rale de Lausanne (EPFL) Institut de math?matiques B?timent MA, Station 8 Tel: +41 21 693 0528 CH-1015 Lausanne Fax: +41 21 693 4303 SWITZERLAND