Network Analysis of miRNA and Cytokine Landscape in Human Hematopoiesis.

The differentiation/maturation trajectories of different blood cell types stemming from a CD34 common ancestor takes place in different biologically relevant multidimensional spaces. Here, we generated microRNA and cytokine profiles from highly purified populations of hematopoietic progenitors/precursors derived from cord blood hematopoietic stem/progenitor cells. MicroRNA and cytokine landscapes were then analyzed to find their mutual relationships under the hypothesis that the highly variable miRNome corresponds to the 'force field' driving the goal of a stable phenotype (here corresponding to the cytokine abundance pattern) typical of each cell kind. The high dimensionality and lack of linearity of the hematopoietic process pushed us to adopt a distance-geometry approach to compare different trajectories, while a complex network analysis was instrumental in revealing the fine structure of microRNA-cytokine relations. Importantly, the approach enabled us to identify a limited number of factors (represented either by microRNAs or cytokines) corresponding to crucial nodes responsible for connecting distinct interaction modules. Subtle changes in 'master nodes', keeping the connections between different regulatory networks, may therefore be crucial in influencing hematopoietic differentiation. These findings highlight the extremely interconnected network structures underlying hematopoiesis regulation and identify key factors in the microRNA/cytokine landscape that may be potentially crucial for influencing network stability.