Efforts to develop culture technologies capable of eliciting robust human blood stem cell growth have met with limited success. Considering that adult stem cell cultures are complex systems, comprising multiple cell types with dynamically changing intracellular signalling environments and cellular compositions, this is not surprising. Typically treated as single-input single-output systems, adult stem cell cultures are better described as complex, non-linear, multiple-input multiple-output systems wherein the proliferation of subpopulations of cells leads to the formation of intercellular endogenously secreted protein interaction networks. Genomic and proteomic tools need to be applied to generate high-throughput (and ideally high-content) biological measurements of stem cell culture evolution. Datasets describing cellular interaction networks need to be integrated into predictive models of in vitro stem cell development. Ultimately, such models will serve as a starting point for the rational design of blood stem cell expansion bioprocesses utilizing dynamic system perturbations to achieve the preferential expansion of target cell populations.