Predicting Protein–protein Association Rates using Coarse-grained Simulation and Machine Learning

Abstract

Protein–protein interactions dominate all major biological processes in living cells. We have developed a new Monte Carlo-based simulation algorithm to study the kinetic process of protein association. We tested our method on a previously used large benchmark set of 49 protein complexes. The predicted rate was overestimated in the benchmark test compared to the experimental results for a group of protein complexes. We hypothesized that this resulted from molecular flexibility at the interface regions of the interacting proteins. After applying a machine learning algorithm with input variables that accounted for both the conformational flexibility and the energetic factor of binding, we successfully identified most of the protein complexes with overestimated association rates and improved our final prediction by using a cross-validation test. This method was then applied to a new independent test set and resulted in a similar prediction accuracy to that obtained using the training set. It has been thought that diffusion-limited protein association is dominated by long-range interactions. Our results provide strong evidence that the conformational flexibility also plays an important role in regulating protein association. Our studies provide new insights into the mechanism of protein association and offer a computationally efficient tool for predicting its rate.

A Multiscale Computational Model for Simulating the Kinetics of Protein Complex Assembly

“…The atomic structure of proteins was reduced to a simplified model in which each residue is represented by two sites [39]. One is the position of its Cα atom, while the other is the representative center of a side-chain selected based on the specific properties of a given amino acid.…”

“…Specific boundary conditions are applied after the diffusions. The new binding conformation is evaluated by either GO-like potential [40], or a coarse-grained physical-based energy functions [39]. The probability of acceptance for this new conformation after diffusion depends on the calculated binding energy.…”

“…In our more recent study, the total energy of binding between two proteins is described by a simple physics-based potential function consisting of three terms [39]. The first component is the electrostatic interaction which was previously used in the Kim-Hummer model [41,42].…”

“… 6 Our previous study demonstrated that our residue-based simulation method can capture the effects of single- and double-point mutations on the association rates [39]. Therefore, the framework of our multi-scale model can be used to study how mutations affect the kinetics of protein complex assembly by applying the same procedure to both wild-type protein complex and to its mutant systems.…”

“…The ionic strength is an adjustable parameter in our residue-based simulation. Our tests showed that the residue-based model can reproduce the effect of the ionic strength on associations [39]. Therefore, by changing the value of ionic strength, our multi-scale method will also be able to evaluate how protein complex assembly can be regulated by solvation effect.…”