search

mummer4 / mummer

Mummer alignment tool
Artistic License 2.0
436 stars 108 forks source link

ANIm calculation for draft genomes #72

Open SilasK opened 5 years ago

SilasK commented 5 years ago

Hello, I have a question, is there a best way to calculate ANI between two (metagenome assembled) draft genomes?

I've seen using mummer -mwith mummer v3. Or take the weighted average over all the nucmer alignments. Which for me makes more sense, as the genomes are fragmented, but however should I filter the alignments, and when yes how with -g -1 -m?

I tried all cases, but the total alignment length was 1.5-2.5 the genome sizes, which doesn't make much sense to me.

gaboentropy commented 5 years ago

While you solve this problem, maybe you'd like to see: https://github.com/ParBLiSS/FastANI

Which is from the Kostas lab: https://www.biorxiv.org/content/early/2017/11/27/225342

Best, -Gabo

P.S. Why "-g -1"? (Rossello-Mora's paper says "default parameters")

SilasK commented 5 years ago

Thank you for the hint