The algorithm uses a fast and exact method; there are no heuristics, so all matches meeting the accept criteria are guaranteed to be found. Alignments are global; all letters of the database sequence must be aligned to a letter in the query sequence. Gaps are not permitted, except for terminal gaps in the query sequence.
Note that it is the longer sequence (genome, chromosome, gene etc.) that is the query; the database contains the short peptides. The name of the command (search_peptidedb) is intended to remind you of this.
Termination options are supported. By default, termination is disabled, equivalent to -maxaccepts 0 -maxrejects 0. In other words, by default the entire database is searched.
Accept options are supported. By default, -maxdiffs 2 is assumed and other accept criteria are not used.
The query file may be in FASTA or FASTQ format.
A database file must be specified using the -db option and must be in FASTA format.
Translated searches are supported for nucleotide queries.
Multithreading is supported.
Standard output files are supported.
Example
usearch -search_peptides proteomes.fa -db peptides.fa \
-userfields out.txt -userfields query+target+qstrand+diffs+tlo+thi+trowdots