proABC-2 (PRediction Of AntiBody Contacts v2) is a web application to predict which residues of an antibody are forming intermolecular contacts with its cognate antigen, as well as the nature of the contacts distinguishing between hydrogen bonds and hydrophobic interactions.
The proABC-2 webserver takes as input the heavy and light chain sequences of an antibody in a FASTA format e.g.:
proABC-2 also allows the user to provide the DNA sequence of the antibody chains which will be translated into the respective protein sequences using the Biopython Seq module.
Please note! proABC-2 only supports the 20 standard amino acids.
The final proABC-2 output are two files: heavy-pred.csv and light-pred.csv respectively for the heavy and the light antibody chains. They consist of several columns:
- Chothia: position of the residue according to the Chothia numbering scheme
- Sequence: residue type
- pt: probability of making a general interaction
- hb: probability of making a H-bond
- hy: probability of making a hydrophobic interaction
The result page also reports a barplot showing, for each residue of the light and the heavy chain, the probability of making a general interaction, a H-bond and a hydrophobic contact.