Cufflinks - Gene Modeling

[KevinM]

Hi everybody,

I used my Cufflinks RNA-Seq data to model new genes and transcripts.
However, after analyzing some examples via IGV, I'm unable to understand why some models have not been extended while a large number of reads are in the 3'UTR.
Does anyone have an explanation about this phenomenon?
Does anyone has experience on de novo modeling by Cufflinks and other models problems ?

Thank you for your help,

Sincerely,

Kevin