gpertea / stringtie

Transcript assembly and quantification for RNA-Seq
MIT License
378 stars 78 forks source link

Odd models #242

Open nhartwic opened 5 years ago

nhartwic commented 5 years ago

This is more of a question than anything. I was hoping someone could tell me why stringtie produces the models it does in this Image.

What you are looking at here in order is...

  1. Output from transdecoder
  2. Output from stringtie.
  3. Intermediate output from my pipeline that doens't matter for this inquiry
  4. Intermediate output from transdecoder.
  5. Alignments of my reads that produced this gtf file

...I get that transcript assembly in low coverage regions is likely to be error prone. I just don't understand why this error occured. What made stringtie want to break up these alignments into two different transcripts?