About SAbPred

SAbPred is a server that hosts antibody computational prediction tools developed by the Oxford Protein Informatics Group (OPIG).

Use the our tools to:

Annotate an antibody variable domain sequence with numbering schemes.
Model an antibody structure via the ABodyBuilder-ML pipeline using ABlooper for loop-modelling.
Predict the side chains of an antibody structure and mutate antibody residues.
Contextualise your antibody's properties against those of clinical-stage therapeutics and human NGS data.
Predict the canonical forms of CDR loops (except CDRH3).
Assess the humanness of your antibody and humanise it using the Hu-mAb protocol.

SAbPred and its tools have been developed by the Oxford Protein Informatics Group (OPIG) in collaboration with UCB, Medimmune, AstraZeneca, and Roche under an open-innovation agreement.

Publications

SAbPred: Dunbar, J. et al (2016). SAbPred: a structure-based antibody prediction server. Nucleic Acids Res., 44, W474-W478. [link]
ABlooper: Abanades, B. et al (2022). ABlooper: fast accurate antibody CDR loop structure prediction with accuracy estimation, Bioinformatics, 38 , 1877-1880.[link]
PEARS: Leem, J. et al (2018). Antibody side chain conformations are position-dependent, Proteins, 86, 383-392. [link]
ANARCI: Dunbar, J. et al (2016). ANARCI: Antigen receptor numbering and receptor classification, Bioinformatics, 32, 298-300. [link]
SCALOP: Wong, W. et al (2018). SCALOP: sequence-based antibody canonical loop structure annotation, Bioinformatics. [link]
TAP: Raybould, M. I. J. et al (2019). Five computational developability guidelines for therapeutic antibody profiling, PNAS, 116, 4025-4030. [link]
Hu-mAb: Marks, C. et al (2021). Humanization of antibodies using a machine learning approach on large-scale repertoire data, Bioinformatics, 37, 4041-4047.[link]

Contact

Feel free to contact us at opig@stats.ox.ac.uk for any issues or misannotations or general enquiries about SAbPred. If you are contacting us about an error you are experiencing with one of our tools, it would be helpful if you could also let us know the job ID, and/or the link to the results page.