I have used AUGUSTUS for gene prediction.
It produces multiple isoforms for some genes. These are listed as XXXX.1, XXXX.2, XXXX.3 etc.
Which isoform can be taken to be the principal isoform i.e. the isoform used in further analysis?
I ask as it seems most genome papers only report functional descriptions for a single isoform.
However, I could report functional descriptions for all isoforms?
- In lieu of this, at least for OrthoMCL clustering the selection of a principal isoform seems paramount (such that the output is not inundated with gene families consisting of matches between different isoforms of the same gene in the same species).
Until now I have simply gone with the isoform labelled ".1" - is this okay?