Genentech/Isosceles

Too many novel isoforms

Closed this issue · 1 comments

Hello,

Thanks for developing this package.

I'm trying to test Isosceles on simulation data using "de_novo_loose" mode. The script I'm using is the same with your example, except the CB tag. However, when summarizing results from no matter se_transcript or gtf exported, it gave me too many novel isoforms (EC/NC/DN). The known isoforms (AP) are only a small fraction of all output isoforms.

Any insights about what's happening here? Do I need to combine some of the novel isoforms based on its starting and end positions?

Oh nvm, the de_novo_loose is designed for more novel isoforms.