The sinaps command predicts an attribute (trait) for a sequence by searching a database of sequences annotated with known values for the attribute. For example, the attribute could be the SSU copy number specified by a copynr= annotation in the database sequence labels. The value of the attribute can be any string of characters allowed as the value of an annotation.
The name of the attribute is specified by the -attr option, which must match the name used in the annotations.
By default, confidence is estimated by boostrapping. If confidence is not needed, this feature can be turned off for faster execution by using the -noboot option. The -boot_subset option (default 32) specifies the number of words in the subset used for bootstrapping and the -boots option (default 100) specifies the number of bootstrap iterations.
The -strand option is required. The value can be plus or both.
Output is written in tabbed text format to the -tabbedout file. Fields are: query label, predicted attribute, bootstrap confidence (number of iterations where this attribute was predicted), strand (+ or -) and the label of the top database hit.
usearch -sinaps seqs.fa -db copynr.udb -attr copynr -tabbedout copynr.txt -strand both