Hi Akila Depending on the distribution of p-values it is totally reasonable to see many P < 0.05 but few Q < 0.05. At times we have seen behavior where Cuffdiff shows slightly too optimistic q-values. I would suggest you try a third method like limma/voom for differential expression analysis and make plots of the most significant results to evaluate whether you think the signal is real or not. Best Jeff On Thu, May 25, 2017 at 1:05 AM akilabioinfo [bioc] < noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User akilabioinfo <https: support.bioconductor.org="" u="" 13107=""/> wrote Question: > P value <0.05 and Q values~ 1 <https: support.bioconductor.org="" p="" 96280=""/>: > > > Hi, > > I would like to compare two groups (control,disease) which includes 3 > replicates in each group. I followed the nature tuxedo protocol > (HISAT,Stringtie and ballgown). In this method, There are ~700 genes showed > p value<0.05 but the Q value ranges very high(0.5<q value="">0.9). But when i > used Tophat-cufflink-cuffdiff protocol, this method gave me reasonable q > value<0.05. Why there is a higher Q value in ball gown? All these genes are > false positive?. Please help me. > > Thanks, > > AkilaRanjith > > > ------------------------------ > > Post tags: ballgown > > You may reply via email or visit P value <0.05 and Q values~ 1 > </q>