Complex Stability is Encoded in Binding Patch Softness: a Monomer-Based Approach to Predict Inter-Subunit Affinity of Protein Dimers

Knowledge about the structure and stability of protein–protein
interactions is vital to decipher the behavior of protein systems.
Prediction of protein complexes’ stability is an interesting
topic in the field of structural biology. There are some promising
published computational approaches that predict the affinity between
subunits of protein dimers using 3D structures of both subunits. In
the current study, we classify protein complexes with experimentally
measured affinities into distinct classes with different mean affinities.
By predicting the mechanical stiffness of the protein binding patch
(PBP) region on a single subunit, we successfully predict the assigned
affinity class of the PBP in the classification step. Now to predict
the experimentally measured affinity between protein monomers in solution,
we just need the 3D structure of the suggested PBP on one subunit
of the proposed dimer. We designed the SEPAS software and have made
the software freely available for academic non-commercial research
purposes at “http://biophysics.ir/affinity”. SEPAS predicts the stability of the intended dimer in a
classwise manner by utilizing the computed mechanical stiffness of
the introduced binding site on one subunit with the minimum accuracy
of 0.72.