Prospective identification of hematopoietic lineage choice by deep learning
Can we predict the fate of a blood stem cell from its morphology and speed in time-lapse microscopy movies? Researchers at the Institute of Computational Biology together with colleagues from ETH Zürich have developed a deep learning algorithm that is trained with millions of image patches of single blood cells. Using the algorithm they can predict the fate of a blood stem cell, that is, the cell type it will differentiation into, up to three generations before conventionally used surface markers.
For more information see the original publication.
Buggenthin, F., Buettner, F., Hoppe, P. S., Endele, M., Kroiss, M., Strasser, M., et al. (2017). Prospective identification of hematopoietic lineage choice by deep learning. Nature Methods, http://doi.org/10.1038/nmeth.4182