Suggestion re: output file naming

[Michael Gooch]

Suggestion to developers: Much of the time when doing analysis of paired
end sequencing, we already have a prefix tag in the filenames that
differentiates the paired1/paired2 files, when LengthSort.pl is run on a
pair, shouldn't the pair2 file take its name prefix from the filename
for the second read file? and have both just tag a ".paired" onto the
end instead of paired1 and paired2? (or at least have a command line
option switch to allow us to select this naming scheme?, maybe have 3
naming schemes, one the way you do it now, another the way i suggested,
and a third that allows us to specify output filenames?) We can always
rename the files but to do it in one step would be more convenient.

[MPC]

Hi,

Yes, it should be possible to do something like this. I guess it
might get a bit confusing with the *.single and *.discard files, which
-- for paired end data -- actually contain both forward and reverse
reads. This was why we originally chose to use the same name prefix
for all four output files.

Best
-Murray