Tuxedo Suite: When To Skip Cufflinks/Cuffmerge?

Question

[jobinv]

The transcripts.gtf file is one of the inputs to cuffdiff. This file could be either:

• the reference transcriptome from, say, iGenomes
• the merged predicted gtf file resulting from a cuffmerge (which in turn takes input from cufflinks)

The first approach can avoid using cufflinks/cuffmerge entirely. I understand that both of these are valid options?

My question is, when following a reference-based pipeline (as opposed to reference-guided or de novo), how does one decide whether it is appropriate to follow a tophat->cuffdiff approach rather than a tophat->cufflinks->cuffmerge->cuffdiff approach? What are the advantages of the former method (besides being less time-consuming), and in particular, what are the disadvantages?

Answer 1

[Sean Davis]

I think it boils down to whether or not you want to discover and quantify transcripts not present in your reference transcriptome. If you want these novel transcripts in your analysis, you need the cufflinks step.

[jobinv]

Do please correct me if I'm wrong about this, but isn't this precisely the difference between a reference-based and a reference-guided pipeline? That in the former, one is not interested in transcripts not present in the reference transcriptome?

[Sean Davis]

You are correct.

[jobinv]

Ok, but then my question still stands: in a reference-based pipeline, is there any reason to include cufflinks/cuffmerge?