Download a command-line version of uniqueProt (Linux and NEW!!: WINDOWS)

read the help pages

UniqueProt  takes a submitted protein list and derives a list of sequence-unique proteins by first comparing the sequences with BLAST and secondly using a greedy algorithm to get a representative set from the BLAST output reaching maximum coverage and minimum redundancy for the representative set. The BLAST-alginment can be circumvented by submitting a pairwise alignment file produced by more accurate alignment programs such as PSI-BLAST or water. The number of built sets/families depends on the HSSP-value, which the program needs as a cutoff parameter (read more about HSSP-values and the theory in the paper to this service).

When using output from uniqueProt please refer to:

Mika, S.; Rost, B.

UniqueProt: creating representative protein-sequence sets.
Nucleic Acids Res. 2003;31(13):3789-3791.

E-mail: mika@cubic.bioc.columbia.edu

pppppppppppppppppppp