search

ParBLiSS / FastANI

Fast Whole-Genome Similarity (ANI) Estimation
Apache License 2.0
355 stars 65 forks source link

parameters for viral genomes #92

Open YiJessePi opened 2 years ago

YiJessePi commented 2 years ago

I was wondering if anyone had the opportunity to test parameters for computing ANI between viral genomes. I guess that the defaults of fragment length of 3000 and minimal number of fragments of 50 is too conservative for small genomes.

boweny920 commented 2 years ago

I too have this question. When i tried running 2 exact same viral contig fna files, the output is none, indicating an ANI<80%, which obviously is incorrect.

YiJessePi commented 2 years ago

It might be because you have less fragments than the minimal number of fragments cutoff. The latest version of fastANI use minimal fraction of fragments instead (and I also recommend to reduce the fragments length a bit)

boweny920 commented 2 years ago

It might be because you have less fragments than the minimal number of fragments cutoff. The latest version of fastANI use minimal fraction of fragments instead (and I also recommend to reduce the fragments length a bit)

This worked for me! Thanks much!