Choosing the ‘--data_type’ parameter for combined PacBio CCS and Nanopore data

ablab / IsoQuant

Transcript discovery and quantification with long RNA reads (Nanopores and PacBio)

Other

128 stars 11 forks source link

Choosing the ‘--data_type’ parameter for combined PacBio CCS and Nanopore data #208

Open biochristmas opened 2 days ago

biochristmas commented 2 days ago

Hi I have two sets of input data, one from Nanopore and the other from PacBio. Could you please advise on the ‘--data_type’ parameter? Should I choose ‘pacbio_ccs’ or ‘nanopore’? My goal is to obtain an annotated file after annotation extension.

andrewprzh commented 2 days ago

Hi @biochristmas

If you want to do in in a single IsoQuant run, I'd suggest to use nanopore as a datatype. However, if PacBio has a noticeably lower coverage, PacBio reads may be ignored in some cases.

Alternatively, you run IsoQuant twice and merge the resulting annotations using, for example, gffcompare.

Best Andrey

biochristmas commented 16 hours ago

Hi @andrewprzh I performed Isoquant twice, once with Nanopore data and once with PacBio data. Now I have two annotation files, and I tried using gffcompare to merge them, but it doesn’t seem quite right; it appears to deduplicate individual annotation files. For the software gffcompare, which is used to merge and deduplicate the results of Isoquant from two annotation files, do you have recommended commands for merging and deduplicating?