There is a model for human cell types
A model for a single-cell accessibility dataset was developed to predict expression in various cell types, including astrocytes. The model was trained on the fetal accessibility and expression atlas. To identify accessible regions specific to each cell type, peak calling results from the original studies of each dataset were used to obtain a union set of peaks.
Cells are decoding the mysteries of the human body
Researchers have developed an atlas of 1.6 million human neural organoid cells, which they claim is the largest atlas of any human cell type. It integrates data from 36 single-cell RNA-sequencing data sets, generated with 26 protocols for producing organoid cells from cultured cells. It was built using three machine-learning algorithms developed by the Human Cell Atlas (HCA) consortium.
The mouse brain has a high-resolution transcriptomic and spatial atlas of cell types
A high-resolution atlas of cell types in the Whole Mouse Brain using Taxonomy Tree and Clustering Methods has been developed. Jaccard-Leiden clustering was used to determine the total number of cells in each cluster. The data were obtained from the CCFv3-based microdissections of the mouse brain and the cells were separated into L4, L3 and L4 clusters.