identified residues were used as docking restraints in HADDOCK.
They generated a large initial population of complexes, and 20% of
the top scoring results were subjected to next iteration, refined and
scored with HADDOCK score,Z-score and electrostatic energy
values. The best cluster of complexes was analyzed with PISA
(Protein Interfaces, Surfaces, and Assemblies) [50].
The study of Rawal et al. used PPD as a complement of experi-
mental study, which allowed for better characterization of the
investigated complex. The in silico part of the work focused on
preparation and examination of a static complex structure. In turn,
a recent work of Sinha et al. investigates dynamic behavior of a
complex of an enzyme (trypsin) and an enzyme-activated protein
(Complement component 4) with PPD and subsequent molecular
dynamics simulations [51]. The docking was performed with Clu-
sPro server. The molecular dynamics step was performed with
Gromacs. Notably, the trajectory analysis was performed with Prin-
cipal Component Analysis which is, next to information theory-
based methods, one of the best ways of getting insight into under-
lying mechanisms of protein function. The study provided some
insights into activation of complement components.
A recent contribution of Prakash et al. can serve as a great
example of a complex study in the field of cancer research, employing
protein–protein dockingmethod [52]. Itfocuses onK-Ras4B. Muta-
tions of this protein are responsible for a significant percent of human
cancers. The protein contains a membrane-anchoring fragment,
which facilitates rejection of the least probable docking results. The
docking was based on previous studies, which provided some insights
into possible dimerization interfaces with probe-based molecular
dynamics. These putative interfaces were used during docking with
RosettaDock. Docking results were grouped in clusters, and cluster
representatives were investigated with regular molecular dynamics
simulations in membrane. This step allowed for rejection of com-
plexes which dissociated, and the complexes which improved their
interaction energy and presented increase in the buried surface area
were chosen for further investigation. The computational part of the
study allowed for identification of key regions crucial for dimeriza-
tion. Subsequent mutagenetic study confirmed the role of postulated
interactions in dimerization. The work of Prakash et al. demonstrates
how the protein–protein docking study, preceded by careful interac-
tion interface prediction and followed by experimental validation can
provide valuable insight into protein behavior.5 Application of Protein–Protein Docking to GPCR Oligomers
G-protein coupled receptors (GPCRs) are a class of receptor pro-
teins that constitute the largest part of transmembrane proteins in
the human genome. Currently they are the molecular target ofProtein-Protein Docking 293