Estimating the Quality of Reprogrammed Cells Using ES Cell Differentiation Expression Patterns

Abstract

Somatic cells can be reprogrammed to a pluripotent state by over-expression of defined factors, and pluripotency has been confirmed by the tetraploid complementation assay. However, especially in human cells, estimating the quality of Induced Pluripotent Stem Cell(iPSC) is still difficult. Here, we present a novel supervised method for the assessment of the quality of iPSCs by estimating the gene expression profile using a 2-D “Differentiation-index coordinate”, which consists of two “developing lines” that reflects the directions of ES cell differentiation and the changes of cell states during differentiation. By applying a novel liner model to describe the differentiation trajectory, we transformed the ES cell differentiation time-course expression profiles to linear “developing lines”; and use these lines to construct the 2-D “Differentiation-index coordinate” of mouse and human. We compared the published gene expression profiles of iPSCs, ESCs and fibroblasts in mouse and human “Differentiation-index coordinate”. Moreover, we defined the Distance index to indicate the qualities of iPS cells, which based on the projection distance of iPSCs-ESCs and iPSCs-fibroblasts. The results indicated that the “Differentiation-index coordinate” can distinguish differentiation states of the different cells types. Furthermore, by applying this method to the analysis of expression profiles in the tetraploid complementation assay, we showed that the Distance index which reflected spatial distributions correlated the pluripotency of iPSCs. We also analyzed the significantly changed gene sets of “developing lines”. The results suggest that the method presented here is not only suitable for the estimation of the quality of iPS cells based on expression profiles, but also is a new approach to analyze time-resolved experimental data.

Terminal differentiation and loss of tumorigenicity of human cancers via pluripotency-based reprogramming

“…Extrapolation of gene expression change between states and the ordering of states to reflect normal biology allows the creation of a roadmap onto which other states may be mapped. We have previously used principal component analysis (PCA) to evaluate degrees of cancer differentiation in comparison to normal differentiating cells or cells representative of various differentiation stages( 24 ); analogous to that recently described for ESCs and iPSCs ( 34 ). The latter analysis hypothesizes that a highly representative component of the PCA would align the samples along a differentiation axis.…”

Bioinformatics Studies on Induced Pluripotent Stem Cell

“…Moreover, the Distance index is defined to indicate the qualities of iPSCs, which based on the projection distance of iPSCs-ESCs and iPSCs-fibroblasts [27]. It consists of two ''developing lines'' that reflects the directions of ESC differentiation and the changes of cell states during differentiation.…”