[Bioperl-l] hmmer3/hmmscan parser

[Thomas Sharpton]

Hi Kai,

I've just pushed the code to github, which you can find here:

http://github.com/bioperl/bioperl-hmmer3

Please use this updated code before making any significant changes - I  
think I may have already fixed the bug you brought up earlier (but  
maybe not?). Do let me know if you have any problems getting ahold of  
this data or if you find any bugs in the code I'd deposited.  Still  
getting my head wrapped around github.

> No worries, that's fine. I've got a checkout of the standalone
> repository that I can play with now. Is there any particular reason  
> you
> decided to create a new parser instead of integrating the code into  
> the
> existing hmmer.pm module? I haven't looked at how the hmmer2 hmmsearch
> output looks compared to the hmmer3 version and if there's any
> conflicts.

Trying to integrate hmmer3 into the old hmmer searchIO module was the  
original idea. But after talking to some of the BioPerl gurus and  
considering the inherent differences between hmmer3 and hmmer2 (at  
least during beta, though there are still some major output report  
differences in the live release), we decided as separate module would  
be ideal.  I don't want to speak out of turn, but it sounds like this  
might be one of the ways that the bioperl project is expanded in the  
future without overbloating bioperl-live. In theory, we can extend  
Bio::Run into this module as well in the future, such that bioperl- 
hmmer3 has a SearchIO path in addition to a Run path.  I don't know  
what the more experienced developers currently think about this idea.

This is an obvious statement, but I feel it's important to be clear on  
these matters - you should feel free to make any and all contributions  
to the development of this module as you see fit.  BioPerl has been  
wonderful to me and I started this module to give a little back, but  
this remains community generated software.

FYI - I have a fix that I'm working on to handle the secondary  
structure track in the alignment report, so if you're particularly  
interested in that data, give me a bit and I'll have it up and running.

All the best,
Tom