repetitive DNA

ckennedy at ems.salk.edu ckennedy at ems.salk.edu
Thu Jan 11 20:11:36 EST 2001


Hello,

I am working on a tiny piece of the Drosophila genome project.
I am trying to assemble regions
from the base of the arms, 
where Celera assemblies stop, and progress towards the centromeres.

The tools i have been provided are Phred,Phrap, Consed, and blast.
My computers are Solaris 7 on sun hardware, Linux 2.2.18 on an iBook,
and Irix 6.5 on SGI Octane.

My project is BACs subcloned into 3kb and 10kb chunks.

One of the problems i have been struggling with is that the region
contains some repetitive DNA elements, some large repeats that
are highly identical, and some of the repeats are larger than my clone
size.   
 
I was thinking about attacking the problem with a divide and defeat
tactic, but i am new to thinking about how to assemble this gnarly
DNA.  What i was thinking of is this:

	1. make some fake projects
        2. copy the files from all the clones i know or strongly 
            suspect overlap into this fake project. 
	3. assemble fake without interference from highly identical 
            sequences some large distance away. 
	4. export the consensus as a high quality 'phd' file
            as a guide for phrap
	5. reassemble the BAC with these guides in place
  
I was thinking of using a similar tactic for 3 and 10 kb clones that 
are obviously placed incorrectly. 

some questions I have :

Am I approaching this entirely wrong ?

Does anyone have any suggestions or hints for working with repeats ?

Are there traditional ways to approach this problem ?

Are there any new assemblers that were designed to address
    repetitive DNA ?

Does anyone know of other tools(websites,programs,books) i may
    find useful ?

Should I drop this project until the tools are available to deal with
    repetitive DNA ?
--

Thank you very much for any help or suggestions.  I have exhausted
nearly all ofmy ideas, and have nobody from whom to recieve peer
review, and discuss this project.   Curt responses pointing me to FAQs,
websites,jounals etc are fine.  

Cameron
----------------------------------------------------







More information about the Comp-bio mailing list