The Human Cell Atlas is a deep learning algorithm method that uses single-cell RNA sequencing to distinguish activated and deactivated cells within humans at any point. Photo by Gio.tto/Shutterstock
Jan. 24 (UPI) -- A new artificial intelligence could help sort normal cells from diseased cells, researchers report in a new study.
The Human Cell Atlas is a deep learning algorithm method that uses single-cell RNA sequencing to distinguish activated and deactivated cells within humans at any point, according to a study published Wednesday in Nature Communications. The ability to pinpoint healthy cells from diseased cells at a given time within a person's life cycle.
"From a methodological point of view, this represents an enormous leap forward. Previously, such data could only be obtained from large groups of cells because the measurements required so much RNA," Maren Büttner, a researcher at the Institute of Computational Biology of the Helmholtz Zentrum München, said in a news release. "So the results were always only the average of all the cells used. Now we're able to get precise data for every single cell."
Researchers from Germany developed the AI system's algorithm based on a large collection of sequences from different single cells.
"The batch effect describes fluctuations between measurements that can occur, for example, if the temperature of the device deviates even slightly or the processing time of the cells changes," Büttner said. "We, therefore, developed a user-friendly, robust and sensitive measure called kBET that quantifies differences between experiments and therefore facilitates the comparison of different correction results."
The team says this method allowed them to apply their results to datasets of millions of cells.
But one researcher warns observers to approach these results with caution.
"We're not developing software to smooth out results. Our chief goal is to identify and correct errors," said Fabian Theis, ICB Director and professor of Mathematical Modeling of Biological Systems at the TUM. "We're able to share these data, which are as accurate as possible, with our colleagues worldwide and compare our results with theirs."