KeyGenes, a Tool to Probe Tissue Differentiation Using a Human Fetal Transcriptional Atlas

•NGS-derived transcriptional profiles of human fetal tissues/organs are generated•Algorithm called KeyGenes uses a training set to predict the identity of a test set•KeyGenes using the fetal atlas identifies NGS- and microarray-derived data•KeyGenes is a flexible and expandable platform to monitor stem cell differentiations

SummaryDifferentiated derivatives of human pluripotent stem cells in culture are generally phenotypically immature compared to their adult counterparts. Their identity is often difficult to determine with certainty because little is known about their human fetal equivalents in vivo. Cellular identity and signaling pathways directing differentiation are usually determined by extrapolating information from either human adult tissue or model organisms, assuming conservation with humans. To resolve this, we generated a collection of human fetal transcriptional profiles at different developmental stages. Moreover, we developed an algorithm, KeyGenes, which uses this dataset to quantify the extent to which next-generation sequencing or microarray data resemble specific cell or tissue types in the human fetus. Using KeyGenes combined with the human fetal atlas, we identified multiple cell and tissue samples unambiguously on a limited set of features. We thus provide a flexible and expandable platform to monitor and evaluate the efficiency of differentiation in vitro.

Graphical AbstractFigure optionsDownload full-size imageDownload as PowerPoint slide