Single-nucleus chromatin accessibility profiling identifies cell types and functional variants contributing to major depression

We performed high-throughput snATAC-seq using the 10X Genomics Chromium platform on archived post-mortem dorsolateral prefrontal cortex (BA9) tissue in MDD subjects who died by suicide and neurotypical control subjects to identify cell-type specific differentially accessible chromatin regions. We further combined snATAC-seq with previously generated snRNA-seq data from the same subjects to generate high-resolution multi-modal accessibility and expression atlas of cortical cells. This identified accessible chromatin regions potentially regulating expression of genes. Further, MDD-associated genetic risk variants were examined for their allele-specific effects on chromatin accessibility and transcription factors binding sites in cell type speciifc manner, elucdiating target risk genes and pathways associated with MDD. Overall design: 10X Genomics Chromium snATAC-seq was performed on nuclei extracted from BA9 of the post-mortem brain tissue of 44 MDD cases who died by suicide and 40 neurotypical controls. We multiplexed nuclei obtained from a pair of male (n=42) and female (n=42) subjects before 10x microfluidic capture and produced combined libraries. These libraries were demultiplexed using common variants and sex-specific chromatin accessibility signals. Unsupervised clustering was perfomed using accessibility measured in 500bp bins and pseudobulk differential accessibility analysis was performed between cases and controls for each cluster and each broad brain cell type. The library preparation was performed with the Chromium Single Cell ATAC v1.1 Chemistry. Paired-end sequencing was performed on the Illumina NovaSeq 6000. FASTQ files were produced with 10x cellranger-atac mkfastq. Alignments were generated with refdata-cellranger-arc-GRCh38-2020-A-2.0.0 and cellranger-atac count. 10x fragment files were converted to peak count matrices using ArchR [release_1.0.1].