Clustered Network Topology of Regulatory System Strengthens the Cellular Robustness to Stress: a Case Study in a Resistant Cultivar of Cassava (Manihot esculenta Crantz) to Viral Infection

Abstract

Robustness tradeoff is one of the essential properties of organisms required to withstand environmental perturbations. It is the balance of robustness and fragility of the biological systems in response to changes of surrounding environment. The property has been studied mostly based on modelling approach whereby the robustness is investigated via the impact of in silico perturbation on the system behaviors. Network topology properties, such as scale-free, small-world, and clustering coefficient, have also been introduced to infer the robustness of biological networks. In this work, we aimed to investigate the topological differences in transcriptional regulatory systems that would bring to the distinct robustness phenotype of an organism. Four gene co-expression networks (GCNs) were constructed to represent the transcriptional regulation of Namikonga (resistant) and Albert (susceptible) cassava cultivars under infection of cassava brown streak virus. The network topology analysis showed no significant differences among the GCNs. However, the distinction was observed in the topology of the group of genes having relatively high node degree. The results showed that the local clustering coefficients representing the average clustering coefficients of such particular genes with same degree in the GCNs of Namikonga were significantly higher than that of Albert. Moreover, GCNs of the susceptible cultivar adapted itself to increase the number of genes with high node degree after infection, whereas the resistant one performed inverse action. Taking all together, we finally proposed that the clustered network structure supports the robustness property of the resistant cultivar. Our findings provided the fundamental property required to improve robustness of the regulatory systems, resulted in resistant phenotypic trait. It would contribute to make a design rationale for crop improvement.