[Biodevelopers] Filtering...

Dan Bolser dmb at mrc-dunn.cam.ac.uk
Thu Aug 14 09:46:31 EDT 2003


Hello, I would like to ask several questions,
and I am sorry if this is the wrong place
to ask. 

What psi-blast filter parameters would
you recomend for DB DB comparison?

Should I prefilter with some dedicated
software and rely on blasts ignorance
of x?

I think my results have loads of repative
domains and coiled coil hits, I remember
that SEG should be tweeked for db vs db,
but I forget how.

Has anyone ever written a PSI-BLAST checkpoint
file parser? Is everything in the .chk available
from the .mat?

Finally, (you will think I am mad), but has
anyone ever implemented blast in perl?

My knowledge of C is basically non existent,
so it makes looking at the (psi-)blast source 
code very difficult. Mostly I never know 'where
we are' in terms of the overall computation.

I always find the best way to understand an 
algorithm is to implement it. I was toying
with the idea of building a perl blast, simply
for that reason - a toy for people to learn
while playing.

Thanks for any thoughts at all,
Dan.





More information about the Biodevelopers mailing list