depression_transcriptome

This code corresponds to:

Systematic review and mega-analysis of the peripheral blood transcriptome in depression implicates dysregulation of lymphoid cells and histones

Authors: Chaitanya Erady, Richard A.I. Bethlehem, Ed Bullmore, Mary-Ellen Lynall

Files:

Differential gene and transcript mega-analyses

• xxx_data_processing.Rmd Data processing instructions for individual datasets
• xxx_DGE.Rmd DGE analysis of individual datasets, Figure S4
• template_for_DTE.Rmd DTE analysis of individual datasets
• meta_analysis_with_bias_and_inflation_correction.Rmd Conducts bias and inflation corrected meta-analysis: DGE cell corrected, DGE not cell corrected, BMI sensitivity analysis, Platelet count sensitivity analysis, DTE cell corrected meta-analysis
• gene_venn.Rmd Creates Venn diagrams, Figure 1A
• qq_plots.r Creates qq plots for meta-analysed results, Figure S5
• LOO_analyses.Rmd Conducts leave-one-out DGE analysis, Data for Table S8
• bacon_weightedz.r Conducts bias and inflation corrected weighted Z meta-analysis
• bacon_weightedz_withNAs.r Conducts bias and inflation corrected weighted Z meta-analysis when data has NA values
• wZ.r Code to conduct weighted Z meta-analysis

Differential transcript usage

• dtu_function.r Functions needed for DTU analysis
• template_for_DTU.Rmd DTU analysis of individual datasets

Comparisons

• compare_DGE_with_proteome.Rmd Compares mega-analytic DGE results to differential protein expression from Daskalakis 2024 Science 384:eadh3707
• RRHO_plots.Rmd Compares DGE to whole blood TWAS results from Meng X et al. 2024 Nat Gen, generates RRHO plots, Figures 1 B,C, and S6 B, C

Core cellular processes mega-analysis

• gene_group_analyses.Rmd Creates data for gene groups correspondign to core cellular processes, permutation test to assess significance, meta-analysis of permutation results
• gene_group_perm_plot.Rmd Creates Figures 2C, S8

Gene/transcript enrichment and cell origin analyses

• GSEA_plots.Rmd Conducts enrichment analysis using Reactome and MSigDB, create leading edge plots, Figures 2A, S7
• gsea_leading_edge.r Code to create leading edge plots
• dge_sc_enrichment.Rmd Performs and plots LR Cell analyses, Figures 2D, S9
• plot_trynka_enrichment.Rmd Makes T cell activation enrichment plot, Figure 2B

WGCNA meta-analysis

• consensus_clustering.Rmd Conducts consensus WGCNA for CNT+MDD samples, Figure 3A,B
• consensus_clustering_control_samples_only.Rmd Conducts consensus WGCNA for CNT samples only
• WGCNA_module_exploration.Rmd Codes for analysis of module enrichment, module preservation scores, module-trait association, module-disorder association, identify hub genes, module membership of DGE genes, inter-dataset correlations, module-smoking association, leave-one-out analysis: without dbGaP, calculate average weighted cohensd, Figures 6A, S10, S11, Data for Tables S4, S5, S6 and S7
• wgcna_cohensd.Rmd Makes WGCNA meta-analysis plots, Figures 3C,D

Processing additional MDD case-control datasets

• To meta-analyse additional MDD case-control datasets, see steps listed for BIODEP/Le/Mostafavi data processing if using RNA-Seq data, else see steps listed for dbGaP/HiTDiP data processing if using microarray data.
• Use meta_analysis_with_bias_and_inflation_correction.Rmd to conduct meta-analysis of processed datasets.
• Use template_for_DTE.Rmd and template_for_DTU.Rmd, respectively, to conduct DTE and DTU analysis.
• Gene-, transcript-level summary statistics and harmonised processed individual-level count matrices along with estimated cell counts and metadata are provided (where permitted) at the Zenodo repository https://doi.org/10.5281/zenodo.15290507.