This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to get Ensembl ids in cuffdiff output?

Hello everyone, I was wondering if there was a way for me to run Cufflinks/Cuffdiff pipeline, so that the output has, besides gene names, the Ensembl id too? (Or Refseq id, Entrez id)

I ran the tophat using Ensembl bowtie2index, "GRCm38_MusMusculus", and I used "gencode.vM12.annotation.gtf" as reference annotation to run Cufflinks and Cuffmerge, but I only get it in my final cuffdiff output (gene_exp.diff), the gene names and XLOC ids, like this: enter image description here

Thank you for help!

cufflinks cuffdiff rna-seq ensembl id
• 4,054 views • link
updated by shunyip shunyip • written by LuisNagano • LuisNagano

1 answer

You can add a gene annotation file (.gtf) that contains Ensembl transcripts into your input, using the -g argument.

• 0 views • link
written by shunyip shunyip

I used (.gtf) of Ensembl and UCSC to run cufflinks and cuffmerge, but it did not work anyway, I just have XLOC ids in my cuffdiff output. Ensembl ids and Nearest_Ref ids only appear on my cufflinks and cuffmerge output, but they disappear on my final output CuffDiff, which I need.

What command line do you use to run cufflinks, cuffmerge and cuffdiff?

My cufflinks output: transcripts.gtf enter image description here

My cuffmerge output: merged.gtf enter image description here

My cuffdiff output: gene_exp.diff enter image description here

My cuffdiff output: genes.fpkm_tracking enter image description here

• 0 views • link
written by LuisNagano • LuisNagano

Cuffmerge assigned new IDs to your genes because they are merged and changed in the output. Because the genes' locations are edited, they are not equivalent to the Ensembl genes anymore, so, Ensembl IDs should not be used in those cases.

Usually, when I use cuffmerge, I simply want to find novel genes. In such a case, I simply use bedtools to find genes that do not overlap with the Ensembl gtf file. So, I never needed to match their IDs.

In your case, maybe you can simply append the novel genes (obtained by bedtools intersect -v argument) onto the ensembl gtf, then use it for cuffdiff?

• 0 views • link
written by shunyip shunyip

Log in to answer this question.

More posts like this