[BiO BB] hat kinds of data mining techniques have been using indrug discovery and drug delivery

Mike Marchywka marchywka at hotmail.com
Tue Jul 17 19:08:03 EDT 2007

I don't really have anything insightful to say directly regarding your 
question but I will point
out that there are good case studies on pubmed- protease and polymerase 
targeting, or
if you prefer kinases, should keep you busy for a while :) As far as 
lots of nano stuff, not sure on tools.

I did want to mention that I have been amazed at the ( apparent) lack of 
some simple
tools however. While it is quite possible I missed some, I have had to write
a lot of scripts and c++ code to augment the Affymetrix annotations. Much of
this is just the novel idea of using a computer to automate data processing 
than require a user to appreciate someone's nice web page for every protein
he wants to investigate. However, then there are things like string 
that execute in reasonable time, programmable ribosomes, format converters 

I suppose I could have looked more carefully at conserved domain servers or
bioperl packages to address various parts of the problem but so far I think
I've done better with my own approach.  I'm not sure I have exploited all
the online tools- I only really use blast and eutils to download batches
of proteins or nucleotides - but I do know that perl, at least under cygwin,
was simply too slow to do anything of any size. I ended up writing my own
c++ string manipulation stuff ( even here under cygwin the STL string
classes were just too slow and I created objects that manipulate c-style
strings ). Even using grep+sed to convert to fasta files was quite slow
( although I think this turned out to be mostly a problem using
cygwin to pipe results - the console buffering seems to be the problem).
I don't want to sound negative on cygwin- you just can't do this with
BAT files and it is hard to get reasonable performance on top of windoze-
but I'm not sure if that is creating some of the limitations.

Don't know if any of that helps but I am curious if anyone has similar or
contrasting observations.


Mike Marchywka
586 Saint James Walk
Marietta GA 30067-7165
404-788-1216 (C)<- leave message
989-348-4796 (P)<- emergency only
marchywka at hotmail.com

>From: "Xue Li" <me.lixue at gmail.com>
>Reply-To: "General Forum at Bioinformatics.Org" 
><bio_bulletin_board at bioinformatics.org>
>To: bio_bulletin_board at bioinformatics.org
>Subject: [BiO BB] hat kinds of data mining techniques have been using 
>indrug discovery and drug delivery
>Date: Tue, 17 Jul 2007 13:33:59 -0500
>Hello all,
>I was wondering what kinds of data mining techniques have been using in 
>discovery and drug delivery? It would be much appreciated if you could 
>me some resources to find it out. Millions of thanks.
>As far as I know, classification techniques are used in protein-protein
>interface prediction, and RNA-  , DNA-  interface prediction.
>Are optimization techniques used? How about regression techniques?
>Li Xue
>Bioinformatics and Computational Biology program @ ISU
>Ames, IA 50010
>General Forum at Bioinformatics.Org - BiO_Bulletin_Board at bioinformatics.org


More information about the BBB mailing list