View Single Post
Old 03-22-2013, 12:52 AM   #4
Daniel1977
Junior Member
 
Location: Spain

Join Date: Mar 2013
Posts: 6
Default

Thanks Kris_143.

Yes one can always use the (a) option.

But if you want to use cuffdiff it seems like a better option to run cuffdiff with all the conditions so that the variance model is more informed, hence you will probably get more precise results.

Also regarding Q4 Iīve tried 0.01, 0.05 and 0.99 and got:

P =0.99
Performed 0 isoform-level transcription difference tests
Performed 0 tss-level transcription difference tests
Performed 38023 gene-level transcription difference tests
Performed 0 CDS-level transcription difference tests

P=0.05
Performed 45266 isoform-level transcription difference tests
Performed 40546 tss-level transcription difference tests
Performed 38023 gene-level transcription difference tests
Performed 39875 CDS-level transcription difference tests

P=0.01
Performed 45266 isoform-level transcription difference tests
Performed 40546 tss-level transcription difference tests
Performed 38025 gene-level transcription difference tests
Performed 39875 CDS-level transcription difference tests

It seems that it isnīt really doing anything (also from other comments in the post), but it puzzles me that the 0.99 option is devoid of all tests but the genes' .
Daniel1977 is offline   Reply With Quote