Can I Use Cummerbund To Explore Replicates Of One Sample Type, And If Not What Should I Use To Explore The Data?

Question

[logothetiskron]

Dear community, I have a dataset that contains three replicate RNA sequencings of a specific cell type. I am very new to this process but have gone through a tutorial from a Broad Institute workshop.

I ran cuffmerge and now have a merged.gtf file from my own data. Typically at this stage people run cuffdiff to look at differential expression. I on the other hand don't want differential expression but rather want to look at the data itself, and the abundance of particular classes of genes. In some cases it helps to compare replicates to themselves, for example to see the sample variance, but what if I want to take an average or get a confidence interval for abundance of a gene or ontogeny-based geneset? For example, a heatmap of the abundance of each member of a protein family (say the rab gtpases).

Can I use cummeRbund for this? Any other suggestions?

Thanks for the help.

Answer 1

[koukougogo]

CummeRbund only takes the output from Cufflinks. After you run Cuffdiff, you can find multiple output files, the *.fpkm_tracking files give you details about the read counts and normalized FPKM value. You can then perform your own statistical analysis. It's just text files, you dont need CummeRbund for this. Hope it helps!