A seed-based approach to identify risk disease sub-networks in human lung cancer

Lung cancer is the leading cause of cancer deaths worldwide. The identification of lung cancer risk disease sub-networks not only helps to understand lung cancer mechanism better, but also provide the potential benefits for the early diagnosis and lead to important applications such as drug targeting. Although some researches are devoted to investigating the carcinogenic process of lung cancer, these approaches have still some limitation. In this paper, the differentially expressed genes are scored and ranked in according to the method of augmented fuzzy measure similarity for obtaining the seed genes. Then, the model of random walk with restarts is used to identify risk disease sub-networks in the PPI network. At last 37 risk disease sub-networks are exploited from the PPI network, which play an important potential role in the carcinogenic process of the lung cancer disease. In terms of the proof and comments in the existing literatures, the identified results show that the proposed method works well in identifying the significant lung cancer risk disease sub-networks, and it is also suitable to recognize other complex risk disease sub-networks.