24 June 2016

New computational method for identifying stem cells

Back to news

Scientists at the Hubrecht Institute and UMC Utrecht have developed an algorithm, called StemID, for the derivation of cell lineage trees and identification of stem cells in single cell mRNA sequencing data. The study is published in Cell Stem Cell this week.

Throughout the life of an organism, mitotic tissues undergo constant turnover. For example the intestine renews itself every 3 to 4 days, and skin every 4 weeks. Key to these processes are our stem cells, capable of dividing and renewing and giving rise to specialized cell types. Knowing the identity of the stem cell is crucial to understanding tissue homeostasis and its aberrations upon disease.

The scientists, led by director of the Hubrecht Institute Alexander van Oudenaarden, have developed an algorithm that can identify stem cells among all detectable cell types within a population using single-cell transcriptomics data.

StemID

The method works by first grouping all similar cells together based on their transcriptome into clusters. These clusters correspond to the cell types present in the adult tissue that is being studied. Then, a lineage tree is computed which infers the developmental connections between all clusters. The stem cells can be then identified by finding the cluster, which has the most connections and the least specialized transcriptome. StemID was successful in identifying adult stem cells in the intestine and the blood. Additionally StemID makes novel predictions about stem cells in the human pancreas.

Finding stem cells

Adult stem cells are very important because they have huge therapeutic potential. As they can self-renew and give rise to several differentiated cell types, they could be used to treat a variety of diseases. In some tissues, for example the small intestine, the adult stem cells are already known. In others, like the pancreas, we have little clues about the presence of stem cells. Finding stem cells in these tissues would be very important for understanding how adult tissue is maintained and could have a big impact for clinical studies.