MEGADOCK: An All-to-All Protein-Protein Interaction Prediction System Using Tertiary Structure Data

Abstract

The elucidation of protein-protein interaction (PPI) networks is important for understanding cellular structure and function and structure-based drug design. However, the development of an effective method to conduct exhaustive PPI screening represents a computational challenge. We have been investigating a protein docking approach based on shape complementarity and physicochemical properties. We describe here the development of the protein-protein docking software package “MEGADOCK” that samples an extremely large number of protein dockings at high speed. MEGADOCK reduces the calculation time required for docking by using several techniques such as a novel scoring function called the real Pairwise Shape Complementarity (rPSC) score. We showed that MEGADOCK is capable of exhaustive PPI screening by completing docking calculations 7.5 times faster than the conventional docking software, ZDOCK, while maintaining an acceptable level of accuracy. When MEGADOCK was applied to a subset of a general benchmark dataset to predict 120 relevant interacting pairs from 120 x 120 = 14,400 combinations of proteins, an F-measure value of 0.231 was obtained. Further, we showed that MEGADOCK can be applied to a large-scale protein-protein interaction-screening problem with accuracy better than random. When our approach is combined with parallel high-performance computing systems, it is now feasible to search and analyze protein-protein interactions while taking into account three-dimensional structures at the interactome scale. MEGADOCK is freely available at .

Megadock-Gpu

“…MEGADOCK [6] is a protein-protein interaction prediction system using FFT-based protein-protein docking by Katchalski-Katzir algorithm [21]. MEGADOCK is implemented by C++, and distributed under the GPLv3 license.…”

“…In recent days, protein-protein docking methods are used not only for prediction of a single protein complex but also for an interactome prediction of proteins in a biological pathway [5,6,7]. Thus, the importance of protein-protein docking in proteomics and systems biology fields has been increased as well as in structural biology field.…”

“…However, ZDOCK unfortunately requires long computation time because it employs seven energy terms for its docking score function calculation. On the other hand, MEGADOCK [6,12] uses more compact energy terms and is 8.8 times faster than ZDOCK 3.0 with comparable prediction accuracy. However, even using MEGADOCK, it is difficult to perform a large-scale interactome prediction because of its computation cost.…”

Untitled

“…To overcome these limitations, bioinformatics methods are expected to be useful for identifying PPIs in functional proteomics. Various computational methods have been proposed to predict PPIs, which include genomic context-based methods [13][14][15][16][17], structure-based methods [18][19][20][21], and sequence-based methods [22][23][24][25][26][27][28][29][30][31][32]. Genomic context-based methods such as gene-cluster and gene-neighbor methods are based on the search of pairs of genes that show a correlated position or behavior; these genes are assumed to encode proteins that interact with each other [13].…”

“…If regions of the target surfaces are similar to the sides of the template interface, these two targets are predicted to interact with each other. Other methods, such as PrePPI [20], predict PPIs by structural alignment combined with secondary structures, while MEGADOCK [21] employs docking simulation to draw inferences on PPIs. The above approaches rely on an abundance of information being available.…”