cluster_tree command

Generates clusters from a tree file by setting an distance threshold (d). A clusters is a maximal sub-tree where the distance from lowest common ancestor node to all leaves in the subtree is <= d.

The distance threshold is specified by the -id option. The value is a fractional identity, so e.g. -id 0.97 means identity 97% or equivalently distance 0.03.

Output is written to a tabbed text file specifed by the -clusterout option. Fields in the file are #1=cluster_number, #2=sequence label.

Example

usearch -cluster_tree seqs.tree -id 0.97 -clusterout clusters.txt