Hi, I have carried out differential expression analysis using Deseq2. One of the most differentially expressed genes is MIR17HG - a host gene for the MIR17-92 cluster, a group of at least six microRNAs
I am wondering if this makes sense? I know micro-RNAs can't be detected in regular RNA-seq but is it correct that a non-protein coding gene is detected?
Thanks,
Zara
Hello,
If non-protein coding genes are included in your reference annotation, then they will be considered by DeSeq2. If you want to avoid those in your analysis, remove them from the reference annotation GTF dataset.
Galaxy tutorials: https://galaxyproject.org/learn/
Thanks! Jen, Galaxy team
Hi Jen,
Thank you for your answer. I do not wish to exclude non-protein coding genes however I have noticed that a large proportion of my most significant differentially expressed genes are non-coding. In your experience does this point to any issue with analysis?
Thanks,
Zara
It may be that you are not using spliced alignment settings during the mapping step.
For example, this is how to set up HISAT2 for spliced mapping (from the Galaxy Tutorials): https://galaxyproject.org/tutorials/nt_rnaseq/#spliced-mapping-with-hisat
