The CDR cluster data download page contains six files including two text files about CDRs, clusters and germlines, a readme file for these text files, and three tar.gz files containing coordinates in mmCIF format. Data can be downloaded via this licensing page.

Commercial licenses can be obtained by emailing Roland Dunbrack (Roland.Dunbrack@fccc.edu).

• File “pyig_cdr_data.txt”: contains one line per CDR, including PDB information, cluster and distance values, sequence, Ramachandran string, germline assignment of the framework and the CDR (which may be different if the antibody is humanized) and their sequence identities, and the sequence of the CDR in the germline.
• File “pyig_domain_data.txt”: contains one line per variable domain, including the same information but for all four CDRs.
• File "pyig_readme.txt": contains descriptions of columns in pyig_cdr_data.txt and pyig_domain_data.txt.
• File “pyig_mmcif.tar.gz”: a tar-gzipped file of mmCIF files for all variable domains in the PDB renumbered according to the modified AHo scheme.
• File “pyig_cluster_mmcif.tar.gz”: a tar-gzipped file of mmCIF files for all clusters separated into separate folders. Each file name includes the name of the cluster for ease of visualization in PyMol or Chimera (so that object names include the cluster identifier, for example: H1-13-1_1H_2J88H_model1.cif
• File “pyig_ab_mmcif.tar.gz”: a tar-gzipped file of mmCIF files for all antibody domain pairs in separate directories (VHVL, VHVH, VLVL) and single domain antibodies (VL, VH), and amyloids.