Research Area: Machine Learning
Computational-intelligence methods in bioinformatics and systems biology show promising potential for leveraging abundant, large-scale molecular data. These methods can facilitate analysis and prediction of the principles of biological systems through the construction of statistical and visualized models. Specifically, structural data from exogenous and endogenous protein-protein interactions are of vital significance in this context, encompassing primarily 3-D structural information for a cohort of macromolecules underpinning the biological system. In this paper, we surveyed the main methodologies and algorithms for the reconstruction and modeling of the structural-interaction networks (SINs) of host- pathogen protein-protein interactions (HPPPIs), regarding how the protein domains interact with each other to constitute a SIN. Surveying the pattern and the organization of the SIN delivers a state-of-theart view of HPPPIs and illustrates prospective future research directions. In addition to the binary PPI network, we distilled the relevant data sources into several branching research areas and further expanded the discussions into computational-intelligence methods according to the algorithms applied, including machine learning statistical models, to shed light on effective method design. In particular, atomic resolution level investigations can reveal novel insights into the underlying principles of the organization and the complexity of HPPPIs networks. Combining data analytics and machine-learning technologies, we anticipate that our systematic overview will serve as a useful guide for interested researchers to carry out related studies on this exciting and challenging research topic in system biology.
Keywords:
Host-Pathogen Protein-Protein Interactions
Bioinformatics
Computational-intelligence
biology
structural-interaction networks
Machine Learning
Deep Learning
Author(s) Name: Huaming Chen; William Guo; Jun Shen; Lei Wang; Jiangning Song
Journal name: IEEE Access
Conferrence name:
Publisher name: IEEE
DOI: 10.1109/ACCESS.2018.2807881
Volume Information: Volume: 6, Page(s): 11760 - 11771
