Wei Liu1, Zhiqiang Sun2, Hongwei Xie2. 1. College of Mechanical & Electronic Engineering and Automatization, National University of Defense Technology, Changsha 410073, China. Electronic address: angel_nudt@126.com. 2. College of Mechanical & Electronic Engineering and Automatization, National University of Defense Technology, Changsha 410073, China.
Abstract
OBJECTIVES: With the announcement of human proteome and interaction data, it becomes possible to comprehensively investigate the tissue-expression and network properties of inherited disease proteins. In this study, our goal was to develop methods to map the disease and expression data and analyze the disorder-tissue associations. METHODS: In this paper, we manually classified the human disease proteins into 22 disorder classes and systematically analyzed the properties of disease proteins in different disorder classes. Then, we investigated the similarity of different disorder classes by computing the overlap of different disorder proteins and networks. We proposed two novel measures, Enrichment Ratio and P-value for comparative analysis of disease proteins across tissues and revealed the associations between disorder classes and tissues/cells. RESULTS: Compared with non-disease proteins, disease proteins tend to express in more tissues, have higher expression levels and interact with more other proteins in the network. The overlap percentages of networks are much higher than those of proteins, implying that different disorder classes usually influence each other by means of their interacting neighbors. The metabolic, muscular and hematologic proteins are related with most tissues/cells, and cancer proteins are closely associated with the disorders in immune cells. CONCLUSION: This paper provided novel methods to investigate proteome-wide disease proteins and their interacting networks in order to understand different disease's associations.
OBJECTIVES: With the announcement of human proteome and interaction data, it becomes possible to comprehensively investigate the tissue-expression and network properties of inherited disease proteins. In this study, our goal was to develop methods to map the disease and expression data and analyze the disorder-tissue associations. METHODS: In this paper, we manually classified the human disease proteins into 22 disorder classes and systematically analyzed the properties of disease proteins in different disorder classes. Then, we investigated the similarity of different disorder classes by computing the overlap of different disorder proteins and networks. We proposed two novel measures, Enrichment Ratio and P-value for comparative analysis of disease proteins across tissues and revealed the associations between disorder classes and tissues/cells. RESULTS: Compared with non-disease proteins, disease proteins tend to express in more tissues, have higher expression levels and interact with more other proteins in the network. The overlap percentages of networks are much higher than those of proteins, implying that different disorder classes usually influence each other by means of their interacting neighbors. The metabolic, muscular and hematologic proteins are related with most tissues/cells, and cancer proteins are closely associated with the disorders in immune cells. CONCLUSION: This paper provided novel methods to investigate proteome-wide disease proteins and their interacting networks in order to understand different disease's associations.