A guide to stem cell identification: Progress and challenges in system-wide predictive testing with complex biomarkers. Novel, systems wide tests can reliably map stem cell phenotypes and potentially replace cumbersome in vivo or in vitro assays. On pages 880-890 Williams, Schuldt and Müller describe new concepts and advances that have lead to the development of a first generation data driven tool for cellular phenotype assessment.
For the graph, human embryonic stem cells (red dots) were differentiated into neural precursor cells (blue dots) and gene expression microarrays hybridized with these samples were nalyzed with PluriTest (www.pluritest.org). The two PluriTest Scores (“Pluripotency Score” and the model-fit assessment score “Novelty Score”) for each sample are plotted on a third dimension representing an empirical density distribution derived from a test dataset with several hundreds of stem cells and somatic samples. Red surfaces values on the PluriTest landscape indicate where pluripotent samples are usually mapped by the algorithm, blue areas indicate empirical scores for somatic samples.
Cover by I. Garitaonandia, J. Goldmann, F.-J. Müller