[mira_talk] Re: Collecting DNA from bacterial cultures, was: suggestion for sequencing companies

  • From: Bastien Chevreux <bach@xxxxxxxxxxxx>
  • To: mira_talk@xxxxxxxxxxxxx
  • Date: Mon, 23 Apr 2012 23:51:57 +0200

On Apr 19, 2012, at 21:57 , Estefania Mancini wrote:
> I was reading carefully last mails about collecting DNA from bacterial 
> cultures and I was wondering how to analyse ORI region in genomes already 
> assembled which resulted in many contigs.
> I've tried with the combination Glimmer + seqinr{oriloc} (R library) but, it 
> is useful for closed genomes in 1 chromosome.
> I usually work with 454 sequences and with bacterial genomes with no 
> reference to align.
> Do you have better ideas for do this?

If I understood you right, you want to find out whether the number of contigs 
is partly due to DNA sample being possibly taken in exponential phase, right?

First I'd check why MIRA stopped building contigs. Look at the ends of the 
contigs in a contig editor. Do they primarily consist of reads with HAF2 tags 
over the endings? Then it might be degraded DNA ... or the 454 sequencing had a 
problem with that bug. In case the contigs end primarily in reads where the 
ends are covered with HAF5, HAF6 and HAF7 tags, then you might be on to 
something.

Without reference it is a bit more complicated I think, but maybe not 
impossible. Note that I have never tried the following, so it may or may not 
work.

I'd try to locate the genes at the origin of replication. Maybe the Glimmer 
annotation is enough, otherwise try to find a couple of more or less close 
relatives in GenBank, extract a couple of protein sequences(!) initiating the 
replication there (dnaA, dnaN and whatever other names were given), then do a 
translated BLAST against the contigs. This should give you a couple of clues 
which contigs (or contig parts) could be candidates.

Then look at the coverage of these contigs, or parts of contigs. If the 
coverage there is consistently >= 1.7x of the average coverage of the whole 
project, then I think you might have a case where the DNA sampling led to 
uneven genome coverage.

Hope this helps,
  Bastien


--
You have received this mail because you are subscribed to the mira_talk mailing 
list. For information on how to subscribe or unsubscribe, please visit 
http://www.chevreux.org/mira_mailinglists.html

Other related posts:

  1. Home
  2. mira_talk
  3. Archive
  4. 04-2012