A Novel Method for Functional Prediction of Proteins from a Protein-Protein Interaction Network

Tae-Ho Kang,여명호,김학용,Jean S. Chung,Jae-Soo Yoo 한국물리학회 2009 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.54 No.4

Functional prediction of unannotated proteins is one of the most important tasks in yeast genomics. Analysis of a protein-protein interaction network leads to a better understanding of the functions of unannotated proteins. Much research has been performed for the functional prediction of unannotated proteins from a protein-protein interaction network. A chi-square method is one of the existing methods for the functional prediction of unannotated proteins from a protein- protein interaction network, but the method does not consider the topology of network. In this paper, we propose a novel method that is able to predict specific molecular functions for unanno- tated proteins from a protein-protein interaction network. To do this, we investigated all protein interaction databases of yeast in public sites such as MIPS, DIP and SGD. For the prediction of unannotated proteins, we employed a modified chi-square measure based on neighborhood counting and we assessed the prediction accuracy of the protein function from a protein-protein interaction network. Functional prediction of unannotated proteins is one of the most important tasks in yeast genomics. Analysis of a protein-protein interaction network leads to a better understanding of the functions of unannotated proteins. Much research has been performed for the functional prediction of unannotated proteins from a protein-protein interaction network. A chi-square method is one of the existing methods for the functional prediction of unannotated proteins from a protein- protein interaction network, but the method does not consider the topology of network. In this paper, we propose a novel method that is able to predict specific molecular functions for unanno- tated proteins from a protein-protein interaction network. To do this, we investigated all protein interaction databases of yeast in public sites such as MIPS, DIP and SGD. For the prediction of unannotated proteins, we employed a modified chi-square measure based on neighborhood counting and we assessed the prediction accuracy of the protein function from a protein-protein interaction network.

Identification of drug target candidates of the swine pathogen Actinobacillus pleuropneumoniae by construction of protein–protein interaction network

Siqi Li,Zhipeng Su,Chengjun Zhang,Zhuofei Xu,Xiaoping Chang,Jiawen Zhu,Ran Xiao,Lu Li,Rui Zhou 한국유전학회 2018 Genes & Genomics Vol.40 No.8

Porcine pleuropneumonia caused by Actinobacillus pleuropneumoniae has led to severe economic losses in the pig industry worldwide. A. pleuropneumoniae displays various levels of antimicrobial resistance, leading to the dire need to identify new drug targets. Protein–protein interaction (PPI) network can aid the identification of drug targets by discovering essential proteins during the life of bacteria. The aim of this study is to identify drug target candidates of A. pleuropneumoniae from essential proteins in PPI network. The homologous protein mapping method (HPM) was utilized to construct A. pleuropneumoniae PPI network. Afterwards, the subnetwork centered with H-NS was selected to verify the PPI network using bacterial two-hybrid assays. Drug target candidates were identified from the hub proteins by analyzing the topology of the network using interaction degree and homologous comparison with the pig proteome. An A. pleuropneumoniae PPI network containing 2737 non-redundant interaction pairs among 533 proteins was constructed. These proteins were distributed in 21 COG functional categories and 28 KEGG metabolic pathways. The A. pleuropneumoniae PPI network was scale free and the similar topological tendencies were found when compared with other bacteria PPI network. Furthermore, 56.3% of the H-NS subnetwork interactions were validated. 57 highly connected proteins (hub proteins) were identified from the A. pleuropneumoniae PPI network. Finally, 9 potential drug targets were identified from the hub proteins, with no homologs in swine. This study provides drug target candidates, which are promising for further investigations to explore lead compounds against A. pleuropneumoniae.

보완된 카이-제곱 기법을 이용한 단백질 기능 예측 기법

강태호(Tae-Ho Kang),유재수(Jae-Soo Yoo),김학용(Hak-Yong Kim) 한국콘텐츠학회 2009 한국콘텐츠학회논문지 Vol.9 No.5

유전체 분석에서 중요한 부분 중 하나는 기능이 알려지지 않은 미지 단백질에 대한 기능 예측이다. 단백질-단백질 상호작용 네트워크를 분석하는 것은 미지 단백질에 대한 기능을 보다 쉽게 예측할 수 있게 한다. 단백질-단백질 상호작용 네트워크로부터 미지 단백질의 기능을 예측하기 위한 다양한 연구들이 시도 되어 왔다. 카이-제곱(Chi-square) 방식은 단백질-단백질 상호작용 네트워크를 통해 기능을 예측하고자 하는 연구 중 대표적인 방식이다. 하지만 카이-제곱 방식은 네트워크의 토폴로지를 반영하지 않아 네트워크 크기에 따라 예측의 정확성이 떨어지는 문제점이 있다. 따라서 본 논문에서는 카이-제곱 방식을 보완하여 정확성을 높인 새로운 기능 예측 방법을 제안한다. 이를 위해 MIPS, DIP 그리고 SGD와 같은 공개된 단백질 상호작용 데이터베이스들로부터 데이터를 수집하여 분석하였다. 그리고 제안된 방식의 우수성을 입증하기 위해 각 데이터베이스들에 대해 카이-제곱방식과 제안하는 보완된 카이-제곱(Modified Chi-square)방식으로 예측해보고 이들의 정확성을 평가하였다. Functional prediction of unannotated proteins is one of the most important tasks in yeast genomics. Analysis of a protein-protein interaction network leads to a better understanding of the functions of unannotated proteins. A number of researches have been performed for the functional prediction of unannotated proteins from a protein-protein interaction network. A chi-square method is one of the existing methods for the functional prediction of unannotated proteins from a protein-protein interaction network. But, the method does not consider the topology of network. In this paper, we propose a novel method that is able to predict specific molecular functions for unannotated proteins from a protein-protein interaction network. To do this, we investigated all protein interaction DBs of yeast in the public sites such as MIPS, DIP, and SGD. For the prediction of unannotated proteins, we employed a modified chi-square measure based on neighborhood counting and we assess the prediction accuracy of protein function from a protein-protein interaction network.

단백질 상호작용 네트워크에서 필수 단백질의 견고성 분석

류제운(Jae Woon Ryu),강태호(Tae-Ho Kang),유재수(Jae-Soo Yoo),김학용(Hak Yong Kim) 한국콘텐츠학회 2008 한국콘텐츠학회논문지 Vol.8 No.6

단백질 상호작용 네트워크는 허브(hub)라 할 수 있는 상호작용 수가 많은 소수의 단백질과 상호작용 수가 적은 다수의 단백질들로 구성된다. 최근 들어 여러 연구들에서 허브 단백질이 비 허브(non-hub) 단백질보다 상호작용 네트워크에 필수적인 단백질일 가능성이 높다고 보고되고 있다. 이러한 현상을 중심-치명 룰(centrality-lethality rule)이라 하는데, 이는 복잡계 네트워크에서 허브단백질의 중요성 및 네트워크 구조의 중요성을 설명하기 위한 방법으로 폭넓게 신뢰받고 있다. 이에 본 논문에서는 중심-치명 룰이 항상 옳게 적용되는지를 확인하기 위해 Uetz, Ito, MIPS, DIP, SGD, BioGRID와 같은 효모에 관한 공개된 모든 단백질 상호작용 데이터베이스들을 분석하였다. 흥미롭게도, 상호작용 데이터가 적은 데이터베이스들(Uetz, Ito, DIP)에서는 중심-치명 룰을 잘 나타냈지만 상호작용 데이터가 대용량인 데이터베이스들 (SGD, BioGRID)에서는 중심-치명 룰이 잘 맞지 않음을 확인하였다. 이에 따라 SGD와 BioGRID 데이터 베이스로 부터 얻은 상호작용 네트워크의 특징을 분석하고 DIP 데이터베이스의 상호작용 네트워크와 비교하였다. Protein interaction network contains a small number of highly connected protein, denoted hub and many destitutely connected proteins. Recently, several studies described that a hub protein is more likely to be essential than a non-hub protein. This phenomenon called as a centrality-lethality rule. This rule is widely credited to exhibit the importance of hub proteins in the complex network and the significance of network architecture as well. To confirm whether the rule is accurate, we investigated all protein interaction DBs of yeast in the public sites such as Uetz, Ito, MIPS, DIP, SGD, and BioGRID. Interestingly, the protein network shows that the rule is correct in lower scale DBs (e.g., Uetz, Ito, and DIP) but is not correct in higher scale DBs (e.g., SGD and BioGRID). We are now analyzing the features of networks obtained from the SGD and BioGRD and comparing those of network from the DIP.

아미노산 - 아미노산 상호작용 네트워크 분석에 의한 단백질의 안정성에 중요하게 영향을 끼치는 아미노산의 파악

유우경,장익수 한국물리학회 2011 새물리 Vol.61 No.9

단백질의 열역학적 및 동역학적 안정성에 중요하게 영향을 미치는 아미노산을 규명하고예측하는 것은 단백질체학에서 중요하다. 단백질의 자유에너지에 근거하여 단백질의 안정성을 연구하는 종래의 방법이여러 가지가 존재하지만, 단백질을 구성하는 아미노산들 사이의상호작용 네트워크를 조명하여 단백질의 안정성의 원천을 규명하는 연구는 드물다. 본 연구에서는 단백질의 구조적 및 에너지적 안정성을 아미노산-아미노산 상호작용의네트워크의 차원에서 규명하기 위한 라플라스 행렬 방법을 제안하였다. CI2 단백질에 대하여분자동역학 시물레이션과 AMBER에너지 함수를 이용하여모든 아미노산-아미노산 짝에 대하여상호작용 에너지를 구한다음 라플라스 행렬 방법을 사용하여CI2 단백질의 열역학적 안정성을 제공하는 아미노산의 세 가지 네트워크 및허브 아미노산들을 규명하였다. 본 연구는 실험적인 연구의 결과들과 함께 비교‧분석이 될 경우단백질 돌연변이체의 디자인 및 열역학적 안정성의 분석 등에 기여할 것으로 여겨진다. In proteomics, identifying and predicting amino acids that affect the thermodynamic and kinetic stability of a protein are important. Although several methods are used to investigate protein stability based on the free energy function of a protein, the same object based on interaction networks of amino acids in a protein structure is limited. Here, we propose a Laplacian matrix method that can be used to systematically investigate the structural and energetic stability of a protein from the aspect of interaction networks of amino acids. For a practical application, we apply this method to the CI2 protein. Then, we clarify that there are three kinds of interaction networks of amino acids, and from each network, we identify a hub amino acid that plays an important role in the thermodynamic and kinetic stability of the CI2 protein. This work,when combined with experimental work, will facilitate the design and the interpretation of the thermodynamic stability of a given protein and is expected to contribute to the development of protein mutagenesis.

Construction of a Protein-Protein Interaction Network for Chronic Myelocytic Leukemia and Pathway Prediction of Molecular Complexes

Zhou, Chao,Teng, Wen-Jing,Yang, Jing,Hu, Zhen-Bo,Wang, Cong-Cong,Qin, Bao-Ning,Lv, Qing-Liang,Liu, Ze-Wang,Sun, Chang-Gang Asian Pacific Journal of Cancer Prevention 2014 Asian Pacific journal of cancer prevention Vol.15 No.13

Background: Chronic myelocytic leukemia is a disease that threatens both adults and children. Great progress has been achieved in treatment but protein-protein interaction networks underlining chronic myelocytic leukemia are less known. Objective: To develop a protein-protein interaction network for chronic myelocytic leukemia based on gene expression and to predict biological pathways underlying molecular complexes in the network. Materials and Methods: Genes involved in chronic myelocytic leukemia were selected from OMIM database. Literature mining was performed by Agilent Literature Search plugin and a protein-protein interaction network of chronic myelocytic leukemia was established by Cytoscape. The molecular complexes in the network were detected by Clusterviz plugin and pathway enrichment of molecular complexes were performed by DAVID online. Results and Discussion: There are seventy-nine chronic myelocytic leukemia genes in the Mendelian Inheritance In Man Database. The protein-protein interaction network of chronic myelocytic leukemia contained 638 nodes, 1830 edges and perhaps 5 molecular complexes. Among them, complex 1 is involved in pathways that are related to cytokine secretion, cytokine-receptor binding, cytokine receptor signaling, while complex 3 is related to biological behavior of tumors which can provide the bioinformatic foundation for further understanding the mechanisms of chronic myelocytic leukemia.

Identifying Responsive Functional Modules from Protein-Protein Interaction Network

Zikai Wu,Xingming Zhao,Luonan Chen 한국분자세포생물학회 2009 Molecules and cells Vol.27 No.3

Proteins interact with each other within a cell, and those interactions give rise to the biological function and dynamical behavior of cellular systems. Generally, the protein interactions are temporal, spatial, or condition dependent in a specific cell, where only a small part of interactions usually take place under certain conditions. Recently, although a large amount of protein interaction data have been collected by high-throughput technologies, the interactions are recorded or summarized under various or different conditions and therefore cannot be directly used to identify signaling pathways or active networks, which are believed to work in specific cells under specific conditions. However, protein interactions activated under specific conditions may give hints to the biological process underlying corresponding phenotypes. In particular, responsive functional modules consist of protein interactions activated under specific conditions can provide insight into the mechanism underlying biological systems, e.g. protein interaction subnetworks found for certain diseases rather than normal conditions may help to discover potential biomarkers. From computational viewpoint, identifying responsive functional modules can be formulated as an optimization problem. Therefore, efficient computational methods for extracting responsive functional modules are strongly demanded due to the NP-hard nature of such a combinatorial problem. In this review, we first report recent advances in development of computational methods for extracting responsive functional modules or active pathways from protein interaction network and microarray data. Then from computational aspect, we discuss remaining obstacles and perspectives for this attractive and challenging topic in the area of systems biology.

Biological Network Evolution Hypothesis Applied to Protein Structural Interactome

Bolser, Dan M.,Park, Jong Hwa Korea Genome Organization 2003 Genomics & informatics Vol.1 No.1

The latest measure of the relative evolutionary age of protein structure families was applied (based on taxonomic diversity) using the protein structural interactome map (PSIMAP). It confirms that, in general, protein domains, which are hubs in this interaction network, are older than protein domains with fewer interaction partners. We apply a hypothesis of 'biological network evolution' to explain the positive correlation between interaction and age. It agrees to the previous suggestions that proteins have acquired an increasing number of interaction partners over time via the stepwise addition of new interactions. This hypothesis is shown to be consistent with the scale-free interaction network topologies proposed by other groups. Closely co-evolved structural interaction and the dynamics of network evolution are used to explain the highly conserved core of protein interaction pathways, which exist across all divisions of life.

Essentiality of Hub Proteins in Protein-protein Interaction Networks of Yeast

Jea Woon Ryu,이윤경,강태호,유재수,정진수,박별나,김학용,여명호 한국물리학회 2010 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.56 No.5

Scale-free protein interaction networks contain a small number of highly connected proteins, called hubs, and a large number of poorly connected proteins. Recently, several independent studies have elucidated that hub proteins are more likely to be essential to cell function than non-hub proteins. Deletion of a hub protein is more likely to be lethal than deletion of a non-hub protein. This concept defines the centrality-lethality rule; it indicates the importance of hub proteins in a complex protein network and the significance of the network architecture. Determination of the link number for a hub protein is obscure. Therefore, it is important to decide how many link numbers the hub proteins have. Here, we propose a new approach for determining the link number of hub proteins. Hub links were counted by locating the intersection point between the power-law distributions of essential and non-essential proteins. Application of this method to the Uetz database yielded an estimate of seven for the minimum number of hub protein links in yeast. Other public database (Ito, DIP,SGD, and BioGRID) predicted a different number of hub protein links. To assess the reliability of the centrality-lethality rule, we examined the essentiality of hub proteins in the protein interaction networks defined within each of the five public datasets: Uetz, Ito, DIP, SGD, and BioGRID. All five sites indicated that hub proteins were more likely to be essential than were non-hub proteins. This new method for determining the number of hub links is a useful tool for hub proteins.

Localized network centrality and essentiality in the yeast–protein interaction network

Park, Keunwan,Kim, Dongsup WILEY-VCH Verlag 2009 Proteomics Vol.9 No.22

<P>It has been suggested that a close relationship exists between gene essentiality and network centrality in protein–protein interaction networks. However, recent studies have reported somewhat conflicting results on this relationship. In this study, we investigated whether essential proteins could be inferred from network centrality alone. In addition, we determined which centrality measures describe the essentiality well. For this analysis, we devised new local centrality measures based on several well-known centrality measures to more precisely describe the connection between network topology and essentiality. We examined two recent yeast protein–protein interaction networks using 40 different centrality measures. We discovered a close relationship between the path-based localized information centrality and gene essentiality, which suggested underlying topological features that represent essentiality. We propose that two important features of the localized information centrality (proper representation of environmental complexity and the consideration of local sub-networks) are the key factors that reveal essentiality. In addition, a random forest classifier showed reasonable performance at classifying essential proteins. Finally, the results of clustering analysis using centrality measures indicate that some network clusters are closely related with both particular biological processes and essentiality, suggesting that functionally related proteins tend to share similar network properties.</P>