Browse COBALT documentation


COBALT does progressive multiple alignment of protein sequences. The alignment is aided by a collection of pairwise constraints derived from conserved domain database, protein motif database, and local sequence similarity using RPS-BLAST, BLASTP, and PHI-BLAST, respectively. Computation time is reduced by forming clusters of sequences that share a large number of common words and finding conserved domains and motif matches for only one sequence per cluster.


Papadopoulos JS and Agarwala R (2007) COBALT: constraint-based alignment tool for multiple protein sequences, Bioinformatics 23:1073-79. PubMed.

Download COBALT software

COBALT software may be downloaded via FTP from