Hi Daniel & Peter, [email protected] schrieb am 12/02/2011 17:58:32:
> Dear Daniel, > > On 12/02/2011 12:07, Daniel Barker wrote: > > A lot of the time for nucleotide stuff it makes sense to search both > > strands. Of course, it isn't hard to search one strand, then the other. > > But this introduces an extra step. I wonder if there could be some > > convenient option to do this, and if it should perhaps be the default? > > (As with NCBI blastall with any kind of nucleotide search.) > > > > This would affect programs beyond just dreg and, though it would be OK > > for our work, perhaps it wouldn't make sense for others. Just a thought. > I think another candidate would be fuzznuc. (That's at least the program, where I sometimes missed this option ;-) > Interesting suggestion. > > Maybe we can add a -bothstrands option for applications to search the > forward and reverse strands. Yes, this would add the new functionality without breaking the old default behaviour of the programs. > We need to consider: > * Do the results make sense? > * What default do we set (maybe some programs have a different default)? As mentioned above, I would not touch the old default settings and add searching both strands as an option. (e.g. in stssearch the search of both strands is already the default.) > * Is this complicated for programs that can use DNA or protein input? > * Can we apply it to applications aligning two sequences? I think it could make sense for programs which can align one sequence against a set of other sequences (e.g. water, needle). Regards, David. _______________________________________________ EMBOSS mailing list [email protected] http://lists.open-bio.org/mailman/listinfo/emboss
