Fig. 1

Experimental approach and integration of snATAC-seq with snRNA-seq clusters. a, Schematic of nuclei isolation and parallel snATAC-seq and snRNA-seq library generation. Temporal cortex tissue was collected from 12 Normal and 12 LOAD donors → Nuclei were extracted from tissue samples by homogenization followed by sucrose gradient purification → Gel beads in emulsion (GEM) generation followed by parallel gene expression and chromatin accessibility library generation according 10X Genomics protocol and sequencing → Cell type annotation and subtype clustering for snRNA-seq data based on gene expression profile comparison to reference dataset, followed by matching of subtype clusters between snRNA-seq and snATAC-seq datasets → Identification of differentially expressed genes (DEGs) and differentially accessible peaks (DAPs) via linear mixed effects modeling→ Identification of cis-coaccessiblity networks (CCANs) incorporating DEGs and DAPs, and categorizing directionality of regulation → Identification of candidate coregulatory elements (cCREs), DAPs coaccessible with DEG regulatory regions, and characterization of biological function category enrichment of associated DEGs → Characterization of enriched TF binding motifs within regulatory regions of DEGs in CCANs → Identification of SNPs impacting TF binding to DEG regulatory regions within 0.5 Mb of GWAS loci. b, Flow chart of snRNA-seq and snATAC-seq analytical pipeline leading to cluster assignment, and schematic of analytical strategy with label transfer of snRNA-seq clusters onto snATAC-seq data by cell type and subtype