Remove miRNA_ref

[lpantano]

Hi all,

we are discussing whether to remove miRNA_ref from the type column and use always isomiRs, and if the reference was used to annotate it, just leave Variant as NA.

Any inputs on this?

[xbdr86]

But, wouldn't that conflict with the gene ontology definition of miRNA_ref and isomiR, since isomiR it's by definition a modification of one or multiple ref_miRNA?

Otherwise, we could go for the parental category of miRNA SO:0000276, but to me that looks like losing a valuable annotation.
http://www.sequenceontology.org/miso/current_svn/term/SO:0000276

[ThomasDesvignes]

That's kind of tricky indeed. Originally we defined the isomiRs as variants of a miRNA_ref so that we could have a base to compare it to and at the same time set, arbitrarily, one isomiR as a standard. I would think that we need to keep this information somewhere.

[lpantano]

Thank for chime in, I think we can leave for now, and se how things go when we have more experience on this.