Sequence length/number requirement

[trickovicmatija]

Hello,
Very lovely and potentially helpful tool!
I am using it to query your database with a relatively small number of microbial genes (~50). The thing is that I don't get any output (there is a header, though, so it's working). I have also manually selected group-specific hashes from many already-created signatures and run mastiff with "--sig". It also doesn't output anything. Do I need a bigger input file because of the short kmer len (21) combined with higher scaling (1000)? What would you say is the optimal number?

P.S. When I run it with a full MAG, I get the output.

Thanks and best,
Matija