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

This deposition accompanies Erady et al. 2025 and includes: - Genome wide mega-analytic results for differential gene expression (DGE) and differential transcript expression (DTE), both sex-pooled and sex-stratified. - Per-dataset differential transcript usage (DTU) results for RNAseq datasets (BIODEP, Le, Mostafavi datasets) - Harmonised processed individual level count data, cell counts and metadata for datasets with permission for public sharing (HitDiP, Le, BIODEP - see below). 1) For files of the format xxx_counts_xxx.tsv: Row names correspond to Ensembl gene IDs and column names correspond to sample IDs Raw gene counts are provided 2) For files of the format xxx_metadata.tsv: Row names correspond to sample IDs 3) For files of the format DGE_MDD_CNT_xxx.tsv Gene names and Ensembl gene IDs are provided. Additional columns correspond to: BacWeightedZ_meta: bias and inflation corrected, and meta-analysed Z-score pvalue.pval_BacWeightedZ: p-value or significance of meta-analysed Z-score pvalue.BacWeightedZ_adj_pval: p-value after Benjamini-Hochberg correction EmpiricalEstimatedMean: bias calculated using the R package bacon EmpiricalEstimatedSD: inflation calculated using the R package bacon 4) For files of the format DTE_MDD_CNT_xxx.tsv Gene names and Ensembl transcript IDs are provided. Additional columns correspond to: BacWeightedZ_meta: bias and inflation corrected, and meta-analysed Z-score pvalue.pval_BacWeightedZ: p-value or significance of meta-analysed Z-score pvalue.BacWeightedZ_adj_pval: Benjamini-Hochberg adjusted meta-analytic p-values EmpiricalEstimatedMean: bias calculated using the R package bacon EmpiricalEstimatedSD: inflation calculated using the R package bacon 5) For files of the format DTU_MDD_CNT_cell_corrected_xxx.tsv Gene names, Ensembl gene IDs and Ensembl transcript IDs are provided. Additional columns correspond to: p_transcript: per-transcript p-value as generated by DEXSeq q_gene: per-gene adjusted p-value - evidence from multiple tests within a gene are aggregated into a single p-value and then FDR corrected for multiple testing across all genes p_transcript_confirmation: stage-wise adjusted p-values - transcript-level p-values corrected for overall FDR using stageR. If the gene-level q-value (q_gene) is >0.05 or the gene has only one transcript, adjusted transcript p-values are not calculated NB. Individual-level count data are also available for the BIODEP dataset, but cannot be deposited publicly according to study ethics - please contact the authors for access; the data are available to all research groups (university email address required). The BIODEP data are available in the same format as detailed above.