Open apcamargo opened 3 years ago
Hi @apcamargo ,
Thanks for the question.
As the readme says, currently Puffaligner does not support spliced-alignment.
The flag --genomicReads
is mainly put for metagenomic alignments. The only difference it has with the decoy option is as follows:
You specify a list of references where the decoy references follow the main ones. Now in enabling the --genomicReads
, in addition to reads that align better to decoy and will be discarded, the reads that align equally good (same score) to decoy and main references will be discarded as well (rather than only reported with the main reference as is in decoy).
Having such an option, in metagenomic analyses, you can have an index over human and microbial sequences all together and listing the human references as decoy to be filtered out. This can replace a two-step pipeline of first filtering the reads that align to (host) human and align the remaining list to a microbial index.
Hey!
I got one question: the
pufferfish align
command has a--genomicReads
parameter, which I assume activates a non-splice alignment mode. However the README says that spliced-alignment is not currently supported.Should we expect
--genomicReads
to give the same results as the default parameters for now? Also, does this mean that you plan puffalign to be splice-aware by default in the future?Thanks!