How does this fit in with models of cell differentiation analysis such as the "Waddington landscape"? The TRA2B upregulated phenotype of cancers would seem to be a "trait" common across all these different tumor niches, as a separate and isolatable component, is that correct?