martin-steinegger
commented
2 years ago MMseqs2 is optimized to process multiple queries at once. So it would make sense to package your search into a big fasta file. If you'd like to perform fast single queries against large databases then our MMseqs2-App (server) might be a good solution. This keeps the index of the target database in memory.
Dear author, I am working a NR annotation for a large number of samples, and I find MMseq2 is a better software. So I would like to ask whether I can run faster together or one by one, My server configuration: CPU 144 and memory 500G. All cat together: 48 G One is ~200 M