Data for ter Kuile et al. 2026 'An integrative Mendelian randomisation and drug mechanism framework for target prioritisation and therapeutic repurposing in major depression': cis-MR results and genetic instruments across 10 QTL datasets

Full cis-Mendelian randomisation (cis-MR) results and genetic instruments for 3,469 druggable genome targets tested across 10 QTL datasets (blood, brain, and cerebrospinal fluid) for association with major depression (MD). These data accompany the manuscript ter Kuile AR et al. (2026) "An integrative Mendelian randomisation and drug mechanism framework for target prioritisation and therapeutic repurposing in major depression", published in Translational Psychiatry (https://doi.org/10.1038/s41398-026-04137-9).

MR results: cis_mr_results_all_targets.tsv

Primary cis-MR results (IVW or Wald) for 3,469 druggable genome targets tested across 10 discovery QTL datasets and 3 replication pQTL datasets against the PGC Phase 3 major depression (MD) GWAS. Includes both significant and non-significant results. Effect estimates are reported as log-odds ratios (beta) and odds ratios with 95% confidence intervals. Fixed-effects IVW was applied for targets with fewer than 4 instruments; random-effects IVW for 4 or more. Bonferroni significance threshold: p < 3.9 x 10-6 (correcting for 12,794 tests across 10 discovery datasets).

Column definitions:
symbol = HGNC gene symbol
ensembl_id = Ensembl gene identifier
uniprot_id = UniProt protein identifier
dataset = QTL dataset name
tissue_type = tissue of origin
modality = molecular trait type (pQTL or eQTL)
analysis_stage = Discovery or Replication
mr_method = Wald (single instrument) or IVW (multiple instruments)
n_instruments = number of genetic instruments after filtering
effect_type = fixed or random effects
beta = log-odds ratio
se = standard error
p_value = p-value
bonferroni_significant = TRUE/FALSE
or = odds ratio
or_lower_ci = OR 95% CI lower bound
or_upper_ci = OR 95% CI upper bound
sample_size_exposure = QTL dataset sample size
sample_size_outcome = MD GWAS sample size
n_cases_outcome = MD GWAS case count

Genetic instruments, files 2-14: [dataset]_variant_table.tsv.gz

Genetic instruments (variant-level data) for the primary cis-MR model in each QTL dataset. One file per dataset. Instruments were selected as cis-acting variants (within 200kb of the target-encoding gene) at p < 1 x 10-6, clumped to LD R-squared < 0.4 using a reference panel of 5,000 UK Biobank participants, with heterogeneity-based pruning applied. Genomic coordinates are on GRCh38.

Column definitions:
symbol = HGNC gene symbol
ensembl_id = Ensembl gene identifier
uniprot_id = UniProt protein identifier
chr_name = chromosome
start_pos = base pair position (GRCh38)
effect_allele = effect allele
other_allele = reference allele
effect_size_exposure = variant-exposure effect estimate
standard_error_exposure = variant-exposure standard error
pvalue_exposure = variant-exposure p-value
f_statistic = single-variant F statistic (z-squared)
effect_size_outcome = variant-outcome effect estimate (log-OR)
standard_error_outcome = variant-outcome standard error
pvalue_outcome = variant-outcome p-value
dataset_exposure = QTL dataset name