Efficient network based approaches for pattern recognition and knowledge discovery from large and heterogeneous datasets

摘要: With rapid technological advances, the potential for transformational science and engineering all scientific domains is enormous. Discovering useful meaningful patterns knowledge extraction from large, diverse, distributed heterogeneous datasets however continues to pose a formidable challenge. Thus, there an urgent need more efficient robust computational approaches effectively manage, use, exploit these data sources. This in turn can accelerate progress of discovery innovation; gain new insights timely manner; lead fields inquiry hitherto impossible. In this dissertation, we tackle challenge by developing applying novel network-based approaches. To demonstrate utility our algorithms, use several large biomedical domain, focusing specifically on rare or orphan diseases (OD) as application. Our research has three facets: First, conduct global network analysis topological analyses deducing underlying biology their causal genes. Specifically, starting with bipartite known OD OD-causing mutant genes, using human protein interactome, functional enrichment literature co-citation, constructed topologically analyzed networks. results revealed that majority disease-causing genes are essential, contrast common which predominantly nonessential.