Analysis of topology properties in different tissues of poplar based on gene co-expression networks

Gene expression analysis is crucial for uncovering components underlying important biological processes for a focal organism. Large-scale gene co-expression networks generally exhibit small-world, scale-free properties; and the degree distributions of these networks follow power-law forms. Topology properties are often informative for determining the key components of the biological systems and their genetic mechanisms. Some basic topology properties display dissimilarity in different tissues, which helps to elucidate the different genetic mechanisms underlying important biological processes among tissues from the top to bottom of trees. In this study, the topology properties of gene co-expression networks were compared in leaf, shoot, wood, and root tissues of poplar. The comparison results demonstrated that the differences of topology properties exist among tissues and the root tissue displays larger average degree and network density, indicating that genes in root tissue are more highly co-expressed than those in the other three tissues. The nodes with a large degree, also known as hub genes, were annotated by the NetAffx Analysis Center and the agriGO tool, with annotation results that these highly interconnected genes are involved in the key biological processes in each tissue. This study also revealed the topology properties’ differences between gene co-expression networks and random network, suggesting the existence of hierarchically organized module and small-world organizations in networks of poplar. The approach described in this research offers an effective strategy for identifying key genes involved in the important biological processes in poplar.