The motifs are expressed as patterns or profiles. The patterns are based on 10-20 amino acid long conserved regions manually selected from multiple sequence alignments, and their diagnostic efficiency has been tested against SwissProt. PROSITE patterns have their own syntax, where [] means alternative and {} forbidden amino acids, and () tells the number of similar amino acids. For example, [IL]-X(2,4)-P-{C} means isoleucine or leucine, 2-4 any amino acids, proline and any amino acid except cysteine. To overcome the diagnostic limitations of patterns, PROSITE also contains profiles, which cover whole domains and cope with gaps and amino acid substitutions, making them more suitable for identifying distant homologues. Each motif contains manual annotation of the structure and function of the protein family or domain in a separate file called PROSITEDOC.

At CSC you can search matches to PROSITE with a protein sequence using the EMBOSS program patmatmotifs. PROSITE home