Our main server is now up and running again!

Please email us if there are any problems.


Sequence data

Access the software



The Abysis database integrates sequence data from Kabat, IMGT and the PDB with structural data from the PDB. It provides a comprehensive point-and-click interface which allows one to search the sequence data on various criteria and display results in different formats. For data from the PDB, sequence searches can be combined with structural constraints. For example, one can ask for all the antibodies with a 10-residue Kabat CDR-L1 with a serine at H23 and an arginine within 10A of H36.


Query the KabatMan database using a point and click interface or using the KabatMan query language. KabatMan is a computer-searchable version of the Kabat antibody sequence data from the Sequences of Immunological Interest book. The KabatMan program is described in the paper: Martin (1996) Accessing the Kabat Antibody Sequence Database by Computer PROTEINS: Structure, Function and Genetics, 25, 130-133.


SeqTest allows you to test your antibody sequence against the Kabat database to identify any unusual features which may result from cloning artifacts or sequencing errors. The sequence you supply is aligned with a consensus sequence in order to apply the standard Kabat numbering and each residue is checked against KabatMan. Unusual residues are reported.

Chothia canonical assignment

Identify the Chothia canonical class assignments for an antibody sequence. When no canonical exists, the program will also identify the most similar class and show which residues cause the mismatch.

Human subgroup assignment

Identify sub-class groupings for human light or heavy chain sequences.

Assess humanness - H-score

Takes an antibody (protein) sequence and compares it with a set of known human sequences. Shows you the similarity Z-score (which we term an H-score) compared with a distribution of scores for human and mouse sequences. This gives you an overall similarity to circulating antibodies.

Assess humanness - G-score

Takes an antibody (protein) sequence and compares it with a sets of known human sequences assigned to germline-derived families. Shows you the similarity Z-score (which we term a G-score) compared with each of these human families. This gives you a similarity to circulating antibodies derived from each human germline family.

Abnum: Apply numbering

Takes a light or heavy chain antibody (protein) sequence and applies Kabat, Chothia, or 'enhanced Chothia' numbering. The server uses a new and very reliable method for applying numbering based on profiles which identify anchor points at the ends of the frameworks. The enhanced Chothia numbering scheme corrects indel sites in framework regions as well as CDRs.

PAPS: Packing Angle Prediction Server

Predicts the VH/VL packing angle and tells you the PDB structure predicted to have the closest VH/VL packing angle to your antibody.