Open jblachly opened 8 years ago
I noticed the total read counts after running in any
mode are extremely low, so what is left over is the detritus of too general of an adapter to look for throughout the read.
That being said, there are still misses, but I think barcodes simply can't be run in -m any
mode due to their length. This is my oversight.
It would be very nice if output could be piped to another stage (or even to another invocation of the program), because a typical workflow could be to remove barcodes (and/or demultiplex while removing), then to remove adapters.
Presently, this requires separate runs, and the way the files are named (with the automatic append of -trimmed) makes this somewhat painful.
When provided with 5 prime adaptors of the form (e.g.)
skewer finds and removes them efficiently in mode
-m head
. However, when running in-m any
mode, they are not detected and removed.