Towards Data Analytics of Pathogen-Host Protein-Protein Interaction: A Survey

Abstract

Abstract"Big Data" is immersed in many disciplines, including computer vision, economics, online resources, bioinformatics and so on. Increasing researches are conducted on data mining and machine learning for uncovering and predicting related domain knowledge. Protein-protein interaction is one of the main areas in bioinformatics as it is the basis of the biological functions. However, most pathogen-host protein-protein interactions, which would be able to reveal much more infectious mechanisms between pathogen and host, are still up for further investigation. Considering a decent feature representation of pathogen-host proteinprotein interactions (PHPPI), currently there is not a well structured database for research purposes, not even for infection mechanism studies for different species of pathogens. In this paper, we will survey the PHPPI researches and construct a public PHPPI dataset by ourselves for future research. It results in an utterly big and imbalanced data set associated with high dimension and large quantity. Several machine learning methodologies are also discussed in this paper to imply possible analytics solutions in near future. This paper contributes to a new, yet challenging, research area in applying data analytic technologies in bioinformatics, by learning and predicting pathogen-host protein-protein interactions. Abstract-"Big Data" is immersed in many disciplines, including computer vision, economics, online resources, bioinformatics and so on. Increasing researches are conducted on data mining and machine learning for uncovering and predicting related domain knowledge. Protein-protein interaction is one of the main areas in bioinformatics as it is the basis of the biological functions. However, most pathogen-host proteinprotein interactions, which would be able to reveal much more infectious mechanisms between pathogen and host, are still up for further investigation. Considering a decent feature representation of pathogen-host protein-protein interactions (PHPPI), currently there is not a well structured database for research purposes, not even for infection mechanism studies for different species of pathogens. In this paper, we will survey the PHPPI researches and construct a public PHPPI dataset by ourselves for future research. It results in an utterly big and imbalanced data set associated with high dimension and large quantity. Several machine learning methodologies are also discussed in this paper to imply possible analytics solutions in near future. This paper contributes to a new, yet challenging, research area in applying data analytic technologies in bioinformatics, by learning and predicting pathogen-host proteinprotein interactions.

Towards Elucidating the Structural Principles of Host-Pathogen Protein-Protein Interaction Networks: A Bioinformatics Survey

“…Given a protein sequence denoted as X = denotes the amino acid types considered in the related projects). In our previous study [9], the protein sequence is decomposed into a 25*n dimension vector. In [18], the utilized database was from PISCES Cull PDB server and the ultimate datasets was filtered based on non-homologous principles.…”

“…Their direct concatenate string becomes the sequence information of proteins. In [9], we give a detailed discussion of the 20 different proteinogenic kinds of amino acids and the sequence information of proteins. However, we have identified that there are 25 different expressions of amino acids existing in the human and pathogens protein sequence information in our PHPPI researches.…”

“…The highly credible positive PHPPIs pairs are manually recorded in these systems and updated periodically. The details of these databases could be found in [9].…”

“…This kind of PPIs offers important information for further analysis of infectious mechanisms between different species. In this paper, we focus on the PPIs between the host and pathogen, in which we benefit from the identified data collected via open databases [9]. These PHPPIs are experimentally verified and manually recorded in systems.…”

“…In one of our recent work [9], a basic sequence information based survey of PHPPI was presented to exploit the online available and experimentally verified PHPPIs data. Beyond classifying pairs of proteins as interacting or not, in this paper we further reach out to a comprehensive study on building structural interaction network for PHPPIs, since systems biology might provide a highly convincing network analysis and also bring trustworthy statistics in cooperation with the corresponding structural information and domain data, on top of the atomic resolution level networks.…”