[Bioperl-l] Issues with new Bio::Tools::Run modules for Genemark and Glimmer

Chris Fields cjfields at uiuc.edu
Fri Mar 2 16:48:57 UTC 2007


On Mar 2, 2007, at 10:15 AM, Mark Johnson wrote:

> On 3/1/07, Chris Fields <cjfields at uiuc.edu> wrote:
> I have been working on an Infernal wrapper (not finished yet but
> getting there) which does this:
>
> Speaking of Infernal...that's on my shopping list, too.  Though we  
> don't invoke cmsearch directly, we use the Sanger rfam_scan wrapper  
> ( http://www.sanger.ac.uk/Software/Rfam/help/scripts/search/ 
> rfam_scan.pl), which does a pre-cmsearch blast to determine which  
> models to run.
>
> I wonder whether to wrap the wrapper, or just incorporate the guts  
> of rfam_scan into a Bioperl wrapper (if the licensing is compatible)?

You could modify it to use both Bio::Tools::Run::StandAloneBlast and  
Bio::Tools::Run::Infernal.  I'm still toying with the latter (only  
cmsearch works) but will probably get back to it this weekend as I  
have a ton of Infernal searches to do.

BTW, the latest Infernal (v0.72, which the wrapper supports) allows  
you to prerun a HMM before running with a full-blown CM search and  
allows you to specify a bit score cutoff to decrease the noise  
level.  Much faster, particularly when running full genome searches...

chris




More information about the Bioperl-l mailing list