Leveraging single-cell ATAC-seq and RNA-seq to identify disease-critical fetal and adult brain cell types

Abstract

Prioritizing disease-critical cell types by integrating genome-wide association studies (GWAS) with functional data is a fundamental goal. Single-cell chromatin accessibility (scATAC-seq) and gene expression (scRNA-seq) have characterized cell types at high resolution, and studies integrating GWAS with scRNA-seq have shown promise, but studies integrating GWAS with scATAC-seq have been limited. Here, we identify disease-critical fetal and adult brain cell types by integrating GWAS summary statistics from 28 brain-related diseases/traits (average N = 298 K) with 3.2 million scATAC-seq and scRNA-seq profiles from 83 cell types. We identified disease-critical fetal (respectively adult) brain cell types for 22 (respectively 23) of 28 traits using scATAC-seq, and for 8 (respectively 17) of 28 traits using scRNA-seq. Significant scATAC-seq enrichments included fetal photoreceptor cells for major depressive disorder, fetal ganglion cells for BMI, fetal astrocytes for ADHD, and adult VGLUT2 excitatory neurons for schizophrenia. Our findings improve our understanding of brain-related diseases/traits and inform future analyses.

Fig. 1. Overview of methods and analyses.

We describe the overview of methods building cell-type annotations from single-cell sequencing datasets (UMAP from) and evaluating disease informativeness applying S-LDSC across GWAS summary statistics. ABC+Roadmap S2G refers to the brain-specific SNPsto-Genes linking strategy using enhancer-gene links^,,,. We separately analyzed fetal and adult brain data.

Fig. 2. Disease enrichments of cell-type annotations derived from fetal brain.

We report A −log₁₀ p-values for positive τ^∗ for a subset of 10 (of 28) diseases/traits and 10 (of 14) fetal brain scATAC-seq cell type annotations; B −log₁₀ p-values for positive τ^∗ for a subset of 10 (of 28) diseases/traits and 10 (of 34) fetal brain scRNA-seq cell type annotations; and C comparison of results for 13 cell types included in both fetal brain scATAC-seq and scRNA-seq data. In A, B, only statistically significant results (FDR > 5%) are colored ( − log₁₀(p-value) ≥ 1.67 for scATAC-seq, ≥ 2.70 for scRNA-seq). In A, B, cell types appearing in both datasets are denoted in red font. Numerical results for all diseases/traits and cell types are reported in Supplementary Data 5, Supplementary Data 7, and Supplementary Data 8. * denotes Bonferroni-significant results. ADHD attention deficit hyperactivity disorder, SCZ schizophrenia, MDD major depressive disorder, BMI body mass index.

Fig. 3. Disease enrichments of cell-type annotations derived from adult brain.

We report A −log₁₀ p-values for positive τ^∗ for a subset of 10 (of 28) diseases/traits and 10 (of 18) adult brain scATAC-seq cell type annotations; B −log₁₀ p-values for positive τ^∗ for a subset of 10 (of 28) diseases/traits and 10 (of 17) adult brain scRNA-seq cell type annotations; C comparison of results for 8 cell types included in both adult brain scATAC-seq and scRNA-seq data. In A, B, only statistically significant results (FDR > 5%) are colored ( − log₁₀(p-value) ≥ 1.79 for scATAC-seq, ≥ 2.04 for scRNA-seq). In A, B, cell types appearing in both datasets are denoted in red font. Numerical results for all diseases/traits and cell types are reported in Supplementary Data 15, Supplementary Data 16, Supplementary Data 17. * denotes Bonferroni-significant results. ADHD attention deficit hyperactivity disorder, SCZ schizophrenia, MDD major depressive disorder, BMI body mass index.

Fig. 4. Comparison between fetal brain scRNA-seq and adult brain scRNA-seq cell-type annotations.

We report A comparison between fetal brain scATAC-seq and adult brain scATAC-seq data and B comparison between fetal brain scRNA-seq and adult brain scRNA-seq cell-type annotations. We report −log₁₀(τ^∗ p-values) of fetal brain scRNA-seq and adult brain scRNA-seq annotations for 6 matched cell types (astrocytes, endothelial cells, microglia, oligodendrocytes, excitatory neurons, inhibitory neurons), conditioning on the baseline model, union of open chromatin regions, and each other. Numeric results are found in Supplementary Data 18 and S19. Correlation among cell-type annotations is found in Supplementary Data 9.