The search_exact command searches for exact, full-length matches to a database sequence. The algorithm is faster than usearch_global, uses less memory, and is guaranteed to find all correct matches, i.e. there are no heuristics.
The query sequences may be in FASTA or FASTQ format.
The database file must be in FASTA format. Indexed (udb) databases are not supported.
The ‑strand option is required for nucleotide databases. If -strand both is specified, then reverse-complemented exact matches will be reported.
Standard output file options are supported. For example, the -notmatched output file can be used to save query sequences that are not found in the database and the -dbnotmatched option can be used to save database sequences that are not present in the query set. This enables incremental updating of dereplicated sequence sets.
Multithreading is supported.
Example
usearch -search_exact seqs.fa -db
db.fa -blast6out matches.b6 -strand plus