Variable domain extraction in antibody-antigen docking


Hi, kindly explain the extraction procedure of the variable domain from the ProABC-2 result. From the tutorial, you stated that “the variable domains of both light and heavy chain represent roughly the first 115-120 residues of each chain”, but the ammino acids from the heavy and light chain of 4G6K pdb is 122 and 109, respectively. Please help, I’m in stock.

my email: aabaukaka@gmail.com
y-abaukaka@razi.tums.ac.ir

Hi,
thanks for your question. A described in the tutorial, ProABC-2 uses the Chothia numbering scheme (https://pubmed.ncbi.nlm.nih.gov/9367782/), which works with insertions and deletions. This is how the numbering and loop identification of antibodies of various sequence length can stay consistent. This means that there can be multiple residues with the same number, followed by a letter. You can see such an insertion on the ProABC-2 result page on residue 31B-G. However we cannot cope with these insertions in HADDOCK, thus the structure must be renumbered to sequential numbering, shifted by the number of insertions or deletions. And you are right that the light chain is even after renumbering shorter than 115 residues, it must have been a typo in the new tutorial. Thanks for letting us know and hope this helps!

Thank you for your reply, I will check the article. Also check the introductory part of HADDOCK 2.4 tutorial where it stated that “i.e. human Fc region should not evoke an immune response in humans.”, in the description of the antibody. I think there is a mistake.

I hope correction is effected as soon as possible. I have to wait for that, because I still can’t get it right.
Thanks.