Question about Delta-Filter for MUMmer

[Brett Bowman]

I'm trying to associate some genomic contigs with one of several references using Mummer. What I can't seem to figure out is how to successfully remove all of the spurious hits.

I align the sequences with nucmer:

nucmer -p <output> <reference> <query>

then remove the short and low-quality hits with delta-filter

delta-filter -l <length> -i <quality>

That gets me ~90% of the way there, but I still have some query contigs aligning to multiple references that I need to remove - primarily those around rDNA genes. I'm getting hits to multiple references for these contigs, but I can't figure out what I'm supposed to do to get rid of them.

Delta-filter has options "-r", "-q", and "-g" which appear to address these sorts of things, but they are not documented nor explained well, and they don't generate the sort of output that I am expect - some pseudo-hits still remain.

Can someone familiar with Nucmer/MUMmer please explain how to filter out the secondary hits from some of my query sequences, or explain what the above options in Delta-Filter are supposed to do?

Sincerely,
Brett

[andylemire]

I've worked a little bit with MUMmer/nucmer/promer, and I typically use delta-filter -1 instead of -r, -q, or -g. This should yield a 1 to 1 mapping of reference and query sequences.