From the revered Tuxedo Suite for NGS data analysis, under what circumstances could the usage of Cuffmerge be thwarted?

Question

[Shaurya Jauhari]

In a scenario where we are to compare paired-end RNA-Seq data from an instance of technical replicate, my understanding is that the Cufflinks output (pointedly transcripts.gtf) from each condition (mock-treated) will represent specific reads; until they are merged together to actually superimpose and render a collective read-gene overlap count matrix. This is the premise for Cuffdiff to unravel differentially expressed genes.

[Devon Ryan]

What is the actual question here? Under what conditions might cuffmerge not work? Given your usage of "thwart", I suspect that something is being lost in translation here.

Answer 1

[colindaven]

Cufflinks is really old and really outdated for differential expression analyses. Why not use something else (edgeR, limma/voom) etc or even Stringtie etc from the same group of authors?