ATLAS maps 40 million mouse brain cells for comparative spatial biology
Published in Bioinformatics, Brain Architecture & Maps, Brain/Neurology, Journals, Tools.
Published October 31, 2026.
ATLAS, a tissue mapping method reported October 30 in Nature Methods, enables large comparative studies of cell distributions in mouse brains.
The method learns transcriptional signatures from single-cell RNA sequencing, encodes them within tissue using oligonucleotide probes, and decodes the measurements to infer cell types and imputed transcriptomes. Those inferred transcriptomes should be distinguished from direct measurement of every gene in each cell.
Researchers profiled 40 million cells across more than 400 coronal sections from 15 mice: five female and five male C57BL/6J animals, plus five male BTBR animals, a model used in autism research. Validation included direct MERFISH measurements of marker genes and comparisons with other technologies.
The analysis identified more than 40 significant differences in cell-type distributions and 16 regional composition changes across the comparisons.
The work supports comparative brain biology. A mouse strain comparison does not establish causes of human autism, and the number of cells does not replace independent animal replication.
Source: Hemminger, Z., Tam, G., De Ocampo, H. et al., Nature Methods, October 30, 2026. Read the original study.