Phenotype-Driven Transitions In Regulatory Network Structure

摘要: Complex traits and diseases like human height or cancer are often not caused by a single mutation genetic variant, but instead arise from multiple factors that together functionally perturb the underlying molecular network. Biological networks known to be highly modular contain dense “communities” of genes carry out cellular processes, these structures change between tissues, during development, in disease. While many methods exist for inferring networks, we lack robust quantifying changes network structure. Here, describe ALPACA (ALtered Partitions Across Community Architectures), method comparing two genome-scale derived different phenotypic states identify condition-specific modules. In simulations, leads more nuanced, sensitive, module discovery than currently available comparison methods. We used compare transcriptional three contexts: angiogenic non-angiogenic subtypes ovarian cancer, fibroblasts expressing transforming viral oncogenes, sexual dimorphism breast tissue. each case, identified modules enriched processes relevant phenotype. For example, specific tumors were associated with blood vessel interferon signaling, flavonoid biosynthesis. structure female male tissue, found has distinct involved estrogen receptor ERK signaling. The functional relevance new indicate only does correlate structural changes, also can such complex networks.