[MUMmer-help] overlapping matches when calling mummer -mum

[Afif E. <ael...@sd...>]

Hello,
I am trying to get a very basic comparison between a de novo assembled 
microbial genome and a reference sequence. I used nucmer at first, but 
there was a tandem duplication that was causing overlapping alignments 
and I just wanted to know precisely what was different. It appears that 
the plain mummer program would be appropriate for finding this out, so I 
called mummer as follows:

mummer -mum <reference> <query>

By using the -mum flag, I thought that I'm not supposed to be seeing any 
overlap in the matches in both the query and reference, but I found an 
instance of this in those positions where I was previously getting 
overlaps in my nucmer alignments:

3569553   3570622    132349
3701734   3702919     39749

If I understood the format of the output correctly, the first match here 
overlaps the second in the reference (3569553+132349 = 3701902 > 
3701734) and the query. By manual inspection of blast results of these 
positions, I know that there is a tandem repeat in this region where the 
query has five copies and the reference has three copies and the overlap 
in the matches here matches all three copies in the reference twice and 
also matches one of the copies in the query twice.

This doesn't look to me like intended behavior of the program, 
especially since I'm not using -maxmatch. Am I doing something wrong or 
misunderstanding anything here? I'm using MUMmer 3.23 as packaged in 
Debian 8. I can provide my query genome privately if necessary.

Many thanks and regards
Afif

-- 
Afif Elghraoui
Laboratory for Pathogenesis of Clinical Drug Resistance and Persistence
San Diego State University
Alvarado Medical Center
6367 Alvarado Court, Suite 206
San Diego, CA 92120
p. 858-222-0454
http://tuberculosis.sdsu.edu