COBALT (6) is a new multiple alignment algorithm that finds a collection of pairwise constraints derived from both the NCBI Conserved Domain database (CDD) and the sequence similarity programs RPS-BLAST, BLASTP and PHI-BLAST. These pairwise constraints are then incorporated into a progressive multiple alignment. COBALT searches can be launched either from a BLASTP result page or from the main COBALT search page (http://www.ncbi.nlm.nih.gov/tools/cobalt/), where either FASTA sequences or accessions (or a combination thereof) may be entered into the query sequence box. A COBALT report will then be displayed with the input protein titles at the top and the multiple alignment at the bottom. From this page, it also possible to get a tree view for the multiple alignment or to launch a modified search using the ‘Edit and Resubmit’ link. In the near future, the tool will provide additional display and download options such as gapped FASTA.
