Network analysis is an extremely powerful tool for reducing the complexity of transcriptomic studies and for identifying key mechanistic information about a biological response. Large-scale meta-analyses utilising network analysis approaches are extremely powerful because of their ability to provide population-level insights that are independent of the inherent biases and limitations of any given study within the larger analysis. While global meta-analyses have been performed on aspects of mononuclear phagocyte biology, previous approaches have generally been limited to a single platform, analysis method, and/or cell population.
Here we provide a comprehensive meta-analysis of inflammatory signalling in mononuclear phagocytes in both mice and men. We generated sub-networks from all previously published studies assessing the transcriptional responses of monocytes, macrophages, and dendritic cells to a wide range of inflammatory stimuli. These sub-networks were then pooled into a single, global network, termed "Mono-Net", which can be used as a basis for identifying and contrasting different aspects of phagocyte biology.
Utilising "Mono-Net", we were able to identify unique sets of mediators that allow for stimulus- and population-level specificity in these populations. We also assessed the suitability of various experimental models for inferring human biology, by comparing inflammatory responses in model systems with those in more physiological patient samples.