Pfam is a protein database, which allows the user to search for functional regions (domains) of a protein sequence. Pfam release version 26.0 is comprised of more than 13K protein families and 499 clans. To successfully locate a protein of interest Pfam utilizes a scoring algorithm similar to other databases such as BLAST, Prosite and SMART. In order to pinpoint a specific domain on the query, a gathering threshold (GA) is assigned to each protein family added by Pfam executives into the database. GA mainly operates under maximum coverage criteria.
Once you have searched a protein sequence, using either the protein code or the accession number using Pfam, you can study the different sequence alignments associated with the sequence, to see how closely related they are and to see if they are significant or not. Furthermore, you can also click on the family links found in the table to find out more information about the domain. For example, you can look at the domain's 3D structure, or you can find out which human proteins contain the specific domain.