What information should I include in "pheno_data.csv" when I do Ballgown analysis?

Question

[lishen0709]

Hello everyone,

I'm now trying to follow the "HISAT2+StringTie+Ballgown" pipline in Nature protocols to do RNA-seq Analysis. In Ballgown analysis, the author of the protocols used a "pheno_data.csv" to give the group information. I'm confused about that. My data is:

ids      groups      reps
ERR1      CCI      rep1
ERR2      CCI      rep2
ERR3      CCI      rep3
ERR4      Sham      rep1
ERR5      Sham      rep2
ERR6      Sham      rep3

What is confusing me is that in their protocols, they had three column (ids, sex, population), So in stattest they set covariate="sex", adjustvars=c("population") and I set covariate="groups", adjustvars=c("reps"). However, there is no DE genes in the final results (all qval > 0.05). So I guess there maybe something wrong with my pheno_data. Could someone help me about this?

I'll be very appreciated if you can help me!