‘Fingerprinting’ cells to speed up analysis

By Nick Carne

Australian researchers have developed a new way of analysing data from individual human cells by combining conventional single-cell techniques with machine learning algorithms.

The resulting process – which they liken to fingerprinting cells – allows single cells with a unique genomic profile to be quickly identified from a tissue sample.

Called scPred, their method provides improvements in the resolution of cell types, they say, and may uncover diseased cells that are outside the scope of current medical diagnostics.

It thus has the potential, they believe, to allow earlier detection of cancer, identify the cells at the root of autoimmune disease, and help personalise treatments to individual patients.

They are now working to translate the method to diagnostic tests for clinical use. 

“For a long time, we’ve mainly classified different cells in the human body based on a limited number of markers found on the cell surface or inside the cell,” says Joseph Powell from the Garvan Institute of Medical Research, the lead author of a paper in the journal Genome Biology. {%recommended 7656%}

“What we’re learning now is that underneath one ‘type’, there is a huge diversity of different cell types. For instance, even though different cancer cells could all have the same cell surface markers, only a subgroup of those cells may actually form a metastatic tumour.” 

scPRed provides a more efficient way of analysing transcripts of individual cells – a measure of which genes are active in different cells, which provides extensive information of what makes cells unique.

The efficiency comes from its ability to collapse all the transcript data from a single cell.

“[I]nstead of trying to estimate 20,000 things at once it works out which patterns of those 20,000 have the most predictive power in distinguishing one cell type from another cell type,” says first author José Alquicira-Hernández, from the University of Queensland.

“scPred then trains a statistical model on those patterns to test what features make a certain cell type ‘most different’ from another cell – which can be thought of as a unique fingerprint.” 

Once a certain cell type has been fingerprinted, the trained model can be used to look for that same cell type in any other sample, in datasets from anywhere in the world.

The researchers say they validated their approach using datasets of colorectal cancer cells analysed by collaborators at Stanford University in the US. Using scPred models, they were able to identify cancer cells from a tissue sample with over 98% accuracy.

“[This method] may allow us to determine the stage of a cancer patient, what potential drugs they will respond to, or whether their tumour cells have signatures that indicate resistance to chemotherapy,” says Powell.