New algorithm makes maps of gene activity easier to compare while preserving cell-level detail
BiologyBiomedical EngineeringData Science
THE AI ANGLE
Unsupervised alignment of spatial gene expression mapsResearchers created an algorithm called Domain Elastic Transform to align spatial transcriptomics tissue maps. It pairs physical cell positions with gene expression to smoothly reshape distorted samples without converting points into pixel grids. This approach helps scientists directly compare cellular organization across developmental stages while preserving single-cell resolution.
THE TEACHING ANGLE
Students can examine the trade-off between converting spatial data into pixel grids, which blurs fine features, and relying on raw coordinates that risk confusing distinct tissue regions with similar shapes.Read the original at phys.org Generate teaching or study materials
More in Biology
- Millions of years of evolution guide AI search for pollution-fighting enzymesPhys.org — Biology · October 3, 2026