ROSMAP participants with repeated neuropsychological assessments and postmortem brain omics (DNA methylation, RNA sequencing, and proteomics) data were included (n=929). Cognitive changes/trajectories were estimated using linear mixed effects (LME)-derived random slopes and group-based trajectory analysis (GBTA)-derived trajectory groups. Single-omics associations (see Single_omics_assoc_.R) were collapsed into gene-level results (see: Add_anno_minP_.R) and meta-analyzed across three omics layers (see: summarize_gene_res.R). NetWAS https://humanbase.net/netwas was then applied using brain tissue–specific networks to prioritize cognitive rates of change or trajectory–related genes, followed by functional enrichment analysis using Database for Annotation, Visualization and Integrated Discovery (DAVID) and g:Profiler https://biit.cs.ut.ee/gprofiler/gost.