Detecting closest reference in custom DB about kmcp HOT 4 CLOSED

Amazing that works like a charm!

from kmcp.

Yes, it does support search reads to detect the most similar strain against many reference genomes. Please follow the recently added tutorial: https://bioinf.shenwei.me/kmcp/tutorial/detecting-pathogens/ .

You may need to remove duplicated sequences first and choose a limited number of representative genomes for each strain. Because, for thousands of reference genomes of small genomes like HBV and dengue viruses, KMCP might fail to detect the target species.

Some tips:

1. Use a larger k value, e.g., 31 or 51, cause you have so many reference genomes and they are highly similar.
2. In the profiling stage, use -n 1 if no targets are detected.

Anyway, just have a try, and let me know if you have any issues.

from kmcp.

Thanks, I'll give it a try. Great tool btw!

from kmcp.

Glad it helps!

I just remembered I forgot to paste the latest unreleased version, which fixes a bug in chunk computation when splitting circular genomes (--circular).

https://github.com/shenwei356/kmcp/files/11702458/kmcp_linux_amd64.tar.gz

from kmcp.