Extended Data Fig. 9: Cell-type Composition Associations, snRNA-seq, and Cell-type Deconvolution. | Nature

Extended Data Fig. 9: Cell-type Composition Associations, snRNA-seq, and Cell-type Deconvolution.

From: Broad transcriptomic dysregulation occurs across the cerebral cortex in ASD

Extended Data Fig. 9: Cell-type Composition Associations, snRNA-seq, and Cell-type Deconvolution.The alternative text for this image may have been generated using AI.

a. ASD bulk RNA-seq log2FC’s (from BA9, left, and BA41-42-22, right) calculated without (x-axis) and with (y-axis) methylation data derived cell-type proportions as covariates (Methods), compared to the original model used for calculating DE genes (Fig. 1). Spearman correlation is shown. b. Split UMAP plots of snRNA-seq from ASD cortex (left) and CTL cortex (right) with cell subtypes coloured. c. Cell-type proportion (mean + standard error) for cell clusters from each region derived directly from the snRNA-seq data (n = 12 biologically independent individuals, 6 per group, sampled across 3 brain regions). Region is coloured and cell-types are indicated in each plot. Nominal and FDR-corrected P values from ANOVA (two sided) are shown for trend-level associations. d. Volcano plots show within cell-type differential expression across regional comparisons. X-axis is ASD effect size (log2FC) and y-axis is the -log10(FDR) from Nebula. Genes with |log2FC| > 0.2 and FDR < 0.001 are labeled.

Back to article page