I've processed a data set that had ribosomal RNA contamination through the evolved tuxedo pipeline (see picture below) after removing rRNA using bbduk from bbmap: [Pipeline for Ribbosomal RNA contaminated reads]
Once I started visualizing volcano plots I noticed some strange artifacts that I haven't been able to put my finger on.
It appears there are several transcripts that are clustering in certain areas of the volcano plots. My first thought was they were NANs and needed removal after means were calculated after ballgown object was created using !=NAN approach.
That did not change the clustering on the volcano plot though, I thought maybe removing exons that = 0 on the string tie output ( CTAB files) may do it but its a bit tricky as all 5 ctab files need equal editing from what I read ?
Any insight is appreciated.