Image Analysis and Classification for High-Throughput Screening of Embryonic Stem Cells

Embryonic Stem Cells (ESCs) are of great interest for providing a resource to generate useful cell types for transplantation or novel therapeutic studies. However, molecular events controlling the unique ability of ESCs to self-renew as pluripotent cells or to differentiate producing somatic progeny have not been fully elucidated yet. In this context, the Colony Forming (CF) assay provides a simple, reliable, broadly applicable, and highly specific functional assay for quantifying undifferentiated pluripotent mouse ESCs (mESCs) with self-renewal potential. In this paper, we discuss first results obtained by developing and using automatic software tools, interfacing image processing modules with machine learning algorithms, for morphological analysis and classification of digital images of mESC colonies grown under standardized assay conditions. We believe that the combined use of CF assay and the software tool should enhance future elucidation of the mechanisms that regulate mESCs propagation, metastability, and early differentiation.