Fig. 1

Brain regions show distinct patterns of sex-differential expression. A PCA plot of all samples and genes before and after removing the effects of SVs. Accounting for SV’s reduces the variance in the data, and cerebellar samples (green) cluster separately from other regions (colors as in C). B SNR from 10,000 iterations, pre- and post-SV adjustment in cortex. SV adjustment reduces the tail of the distribution, makes the distribution more normal, and reduces mean SNR. C SV-adjusted SNR values (10,000 iterations). Each distribution is significantly different from its corresponding null distribution by Wilcox test. All regions other than substantia nigra have a mean SNR value greater than one, providing evidence that there is an expression difference between males and females in multiple brain regions. Cortex has the third highest mean SNR value and has the shortest tail, suggesting that its SNR is highly repeatable. D Summary of DE gene count per region, including omnibus. There is an abundance of DE autosomal genes in the nucleus accumbens, cortex, cerebellum, putamen, and caudate. Sample number does not fully explain the number of DE genes in a given region. E Volcano plot highlighting that allosomal genes follow expected trends. F Volcano plot highlighting the autosomal genes, including noteworthy long non-coding RNA LINC01597. G PCA of all LogFCs from all regions and omnibus, shows omnibus truly represents the average sex-differential expression across all brain regions. H PCA of top 500 most variable LogFCs that were FDR significant in at least one of the “sex-differential” regions. Highlights the fact that cerebellum remains an outlier even when only considering sex differences. I PCA plots showing the key genes that separate the SDR in the same PCA space as panel H. Highlights a few notable genes that can be used to distinguish regions from a sex-differential lens. J Correlation of LogFCs between CMC, and GTEx analysis demonstrates results are robustly shared for the cortex. K Correlation of LogFCs for DE genes between CMC and GTEx shows a high replicability in cortex