Search
Question: Ballgown for differential expressed genes
0
6 weeks ago by
marasodi0
marasodi0 wrote:

I have differential expression RNA seq data, I need to run it on ball gown, the data is between two genotypes, at 4 time points each treated sample have its control and both the control and the treated plant are in triplicates, I used this

pData(bg_chrX_filt) = data.frame(id=sampleNames(bg_chrX_filt), group=rep(c(1,0), each=24)) when I run stattest

results_transcripts = stattest(bg_chrX_filt, feature="transcript", covariate="Time", getFC=TRUE, meas="FPKM")

I get this error

Error in stattest(bg_chrX_filt, feature = "transcript", covariate = "Time",  :

invalid covariate name

modified 6 weeks ago by James W. MacDonald46k • written 6 weeks ago by marasodi0
0
6 weeks ago by
United States
James W. MacDonald46k wrote:

The help page for stattest says this:

timecourse: if 'TRUE', tests whether or not the expression profiles of
genomic features vary over time (or another continuous
covariate) in the study.  Default 'FALSE'.  Natural splines
are used to fit time profiles, so you must have more
timepoints than degrees of freedom used to fit the splines.
The default df is 4.

covariate: string representing the name of the covariate of interest
for the differential expression tests.  Must correspond to
the name of a column of 'pData(gown)'. If 'timecourse=TRUE',
this should be the study's time variable.

If you are doing a timecourse study, you have to tell stattest that's what you are doing. And if there is no 'Time' variable in the pData object for your gown object, then stattest is going to tell you that it can't find it.