Re: [MUMmer-help] MUMmer's issue with overlapping alignments (request for help)

[Adam P. <aph...@gm...>]

Hi,
Yes, that's a known bug in how nucmer extends alignments. There is a
workaround in the latest release. You can run it like so:

> nucmer --maxmatch --banded -D 10 ref.fasta qry.fasta

This will force nucmer to break an alignment at an indel >10 bp. You
can adjust the -D value to the size of an indel you want included in
the alignment. Although setting it too high is pointless, because
alignments will start terminating due to poor alignment scores.

I believe this should fix the problem. Let me know if it doesn't.

Best,
-Adam


On Thu, Jan 24, 2013 at 7:13 PM, PLD <pla...@gm...> wrote:
> Dear all,
> May I ask for your advice please?
>
> At one point, when I was aligning a reference (NC_010079) to a set of
> contigs, MUMmer recognizes two overlapping alignments:
> 1873926    1902489    65901    37311    28564    28591    99.90    2872915
> 65901    0.99    43.38    NC_010079    contig_64
> 1882538    1939799    57289    1    57262    57289    99.93    2872915
> 65901    1.99    86.93    NC_010079    contig_64
>
> And consequently, dnadiff marks this as GAP:
> contig_64    GAP    57290    37310    -19979    -19952    -27
>
> As this is not what I expected, I tried checking the same region using
> Blast.
> I found that Blast saw this case as a single alignment (instead of two
> overlapping ones):
> contig_64    NC_010079    99.92    65902    26    16    1    65901
> 1939799    1873926    0    1.21E+05
>
> Could this be one example of a known problem that you described on the
> MUMmer 3 manual?
> "nucmer, promer and run-mummer3 all have a difficult time with tandem
> repeats. If the two sequences contain a different number of copies of the
> same tandem repeat, these alignment routines will sometimes generate a
> cluster on either side of the tandem and extend alignments past one another,
> failing to join them into a single alignment region. This generates two
> overlapping alignments and makes it difficult to determine what caused this
> erratic behavior."
> (http://newmbcrr.dfci.harvard.edu/mummer/index.html#nucmer)
>
> If so, may I know if you have advice on fixing this please? Is there any
> parameter that I can change to make MUMmer's results in this case resembles
> Blast's results above?
>
> I would really appreciate your help. Thank you very, very much for your
> time!
>
> ------------------------------------------------------------------------------
> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
> MVPs and experts. ON SALE this month only -- learn more at:
> http://p.sf.net/sfu/learnnow-d2d
> _______________________________________________
> MUMmer-help mailing list
> MUM...@li...
> https://lists.sourceforge.net/lists/listinfo/mummer-help
>