Hello -

There are many approaches you can use to try and find a promoter with 
variable degrees of success. There is extensive literature on this. You 
could make profile HMMs and train them with real examples. You could also 
use a Gibbs Sampler. There are examples of both in the biojava in anger 
pages http://www.biojava.org/docs/bj_in_anger/

Other approaches would be programs like MEME or the technique called 
nested MICA developed by Thomas Down of biojava fame which seems to be 
very good.

- Mark





Ilueny Santos <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
12/16/2005 04:35 PM

 
        To:     biojava-l@biojava.org
        cc:     (bcc: Mark Schreiber/GP/Novartis)
        Subject:        [Biojava-l] Locating promoter regions in sequence of 
DNA with   Biojava


Hello to all, 
 
 First would like to be thankful all, in special to the Mark and Gregory, 
for having answered. 
 
 Explaining of form more detailed my doubt: 
      I was trying to locate definitive regions in a DNA sequence (-10 box 
and -35 box).  These regions are small stretches of 6 pairs of bases (pb) 
and are thus called by being generally the 10 pb and 35 pb, respectively, 
upstream of +1 (ATG) and the presence of them in the sequence strong 
characterizes the existence of a promoter. 
 
     The problem is that they are not steady, for example:
 region -10 box normally is presented as TATAAT but it can have variations 
in form TATAAG in such a way or TATTAT how much in its positioning in 
relation to start codon (+1 ATG) 
 
 
 Leaving of Displayed I ask: 
 it will be that I obtain, using biojava it, to make one algorítmo capable 
to locate unstable regions (in such a way in the form how much in its 
positioning) in DNA sequences? 
 
 
 I am thankful all one more time that will be able to help. 
 
 PS.:  Gregory favours, already I am studying Regular Expressions and..., 
Mark, the bayesiano classifier already is fact, but, followed its tip, 
 I go to also study the package org.biojava.dist because it can be useful 
of some form, thanks.
 

 
---------------------------------
 Yahoo! doce lar. Faça do Yahoo! sua homepage.
_______________________________________________
Biojava-l mailing list  -  Biojava-l@biojava.org
http://biojava.org/mailman/listinfo/biojava-l




_______________________________________________
Biojava-l mailing list  -  Biojava-l@biojava.org
http://biojava.org/mailman/listinfo/biojava-l

Reply via email to