Fig. 3: Transcriptional signatures of neuronal and glial populations, and cellular compositional changes in the prenatal Down syndrome brain.

a–f Volcano plots depicting differentially expressed genes (DEGs) in a newborn layer II-III excitatory neurons (NB L2-3 ExN), b newborn layer II-IV excitatory neurons (NB L2-4 ExN), c general interneurons (InN 1), d LHX6-expressing interneurons (InN 2), e astrocyte progenitors (Astro P), and f microglia (Micro) from the prenatal human single-nucleus (sn)RNA-seq dataset. Each dot represents a gene and dots are colored according to enrichment: significantly upregulated (log₂FC > 0, FDR ≤ 0.05) in teal, significantly downregulated (log₂FC < 0, FDR ≤ 0.05) in purple, and non-significant genes in gray. Comparison is between DS and euploid samples. The horizontal line depicts FDR = 0.05, and the vertical line depicts log₂FC = 0. g, h Microenvironment analysis of the Down syndrome (DS) (g) ventricular zone (VZ) and h cortex, from Slide-seq, where a central cell type is selected, and the proportion of the top 100 nearest cells relative to this central cell (y-axis) is calculated in comparison to other cell types of interest (x-axis) and across conditions. Teal indicates an increased abundance of query cells near the central cell in DS compared to control, while purple indicates a decreased abundance. Asterisks (*) denote a false discovery rate (FDR) ≤ 0.05. i, j Leave-one-subject-out (LOSO) analysis across brain region: i VZ, j cortex. In the LOSO approach, the microenvironmental analysis was iteratively performed while excluding one sample at a time. Mean effect sizes and standard deviations across iterations were used to assess the reproducibility of changes. Teal represents consistently upregulated interactions, purple indicates consistently downregulated interactions, and gray indicates variability in the direction of change across iterations.