Hello I have a questions about the rna seq pipeline(Differential expression of genes). some features are in bacteria genome that don’t allow us to get proper result from Eukaryotes pipeline. What are they? I know that 15% or more of bacterial genes overlap other genes what is EDGE-pro says. Are There other reasons? Thanks a lot
•
written
by
zh.khodadadi •
Hi zh.khodadadi,
an educated guess would be that Eukaryotes have spliced transcripts and the pipelines are designed to find splice sides in the data. Some aligners might run into troubles having no splice sites detected.
Cheers,
Michael
thanks for answer. I know but are you run these pipeline( such as Tuxedo) on Prokaryotes? (that Guide me about some troubles and defult parameters require change for instance tophat2 or cufflinks)
You can exchange Tophat2 with Bowtie2 - the index is the same but Bowtie2 does not look for junctions.
Cufflinks might work (it also is designed for several transcripts per gene), but you could have it faster/easier with e.g. HTseq-count or featureCount.
thank for all cooperation in this subject
Hi zh.khodadadi,
This is the third thread in which you want to use the tuxedo pipeline for prokaryotic RNA-seq. The answers in each previous thread (EDGE-pro: Estimated Degree of Gene Expression in Prokaryotic Genomes & Prokaryotic Differential expression analysis- RNA seq data and software ) have been the same: "yes you can but why would you because it's unnecessary and not optimal".
So if you really want to and won't listen to what people here say about it, sure, go ahead.
Log in to answer this question.
