13 · Manuscript Plots, Tables & Data Sharing

Overview

This final pipeline generates all publication-ready figures, summary tables, and data-sharing files from the upstream analysis results. It produces the eQTL and SMR manuscript tables, QC and results plots, and reformatted eQTL summary statistics packaged for public data deposition. Most rules operate independently and can be re-run selectively as manuscript revisions require.

Workflow Logic

graph TD
    classDef snazzy fill:#f1f1f1,stroke:#333,stroke-width:2px,color:#000;
    classDef out fill:#e1f5fe,stroke:#01579b,color:#000;

    A[TensorQTL perm output] --> B(eqtl_tbl)
    C[SMR results] --> D(smr_tbl)
    A --> E(eqtl_qc_plt)
    F[S-LDSR results] --> G(ldsr_plt)
    H[Replication results] --> I(replication_plt)
    J[Genotype PCA covariates] --> K(supplementary_plt)
    A --> L(data_perm)
    M[SMR nominal input] --> N(data_nominal)
    L --> O(mk_eqtl_tar)
    N --> O
    B --> P[(results/13MANUSCRIPT_PLOTS_TABLES)]
    D --> P
    E --> P
    G --> P
    I --> P
    K --> P
    O --> P

    class A,B,C,D,E,F,G,H,I,J,K,L,M,N,O snazzy
    class P out

  1. eqtl_tbl: Generates the main eQTL results table for the manuscript from TensorQTL permutation output, annotating lead SNPs with allele information and overlap with Ziffra 2021 ATAC-seq peaks.
  2. smr_tbl: Compiles the SMR/HEIDI results table across all cell types and GWAS traits, merging SMR output with eQTL nominal summary statistics.
  3. eqtl_qc_plt: Produces eQTL QC figures including signal summaries per cell type and comparisons with the Ziffra 2021 dataset.
  4. ldsr_plt: Plots S-LDSR stratified heritability enrichments across cell types and neuropsychiatric traits.
  5. replication_plt: Produces π₁-based replication plots against published datasets (Bryois, O’Brien, Wen, Fugita) and internal replication results, including beta-correlation scatter plots.
  6. supplementary_plt: Generates supplementary figures, including genotype PCA plots and expression PC summaries across cell types.
  7. data_perm: Reformats TensorQTL permutation-pass eQTL results into a standardised, publicly shareable TSV (annotated with allele information from the PVAR file). Outputs a representative NPC-2 file as the data-sharing example.
  8. data_nominal: Reformats TensorQTL nominal-pass summary statistics from the SMR input directory into a publicly shareable TSV.
  9. mk_eqtl_tar: Packages all per-cell-type permutation and nominal eQTL summary statistic files into a single compressed tar archive (nbray_2026_dev_brain_eqtl.tar.gz) for data deposition.

Output Structure

All outputs are written to results/13MANUSCRIPT_PLOTS_TABLES/:

results/13MANUSCRIPT_PLOTS_TABLES/
├── plots/
│   ├── eqtl_qc_plt.pdf
│   ├── ldsr_plt.pdf
│   ├── replication_plt.pdf
│   └── expression_PC_plt.pdf
├── tables/
│   ├── all_eqtl_tbl.xlsx
│   └── all_smr_tbl.xlsx
└── data_sharing/
    ├── NPC-2_cis_eQTL_perm.tsv.gz
    ├── NPC-2_cis_eQTL_nominal.tsv.gz
    └── nbray_2026_dev_brain_eqtl.tar.gz

Technical Requirements

Category Detail
Container r_eqtl.sif (all R rules)
Modules pigz (mk_eqtl_tar, data_weights)
Input TensorQTL perm/nominal output (05); S-LDSR results (09); SMR results (10); replication results (06)
Output PDF figures, XLSX tables, compressed TSV summary statistics

Resource Profile

Rule Threads RAM Walltime
eqtl_tbl 1 20 GB 2h
smr_tbl 10 80 GB 2h
eqtl_qc_plt 1 10 GB 1h
ldsr_plt 1 10 GB 1h
replication_plt 1 10 GB 1h
supplementary_plt 4 20 GB 1h
data_perm 1 10 GB 1h
data_nominal 10 80 GB 4h
mk_eqtl_tar 8 16 GB 5h
Note

Selective execution

Most rules in this pipeline have no dependencies on each other and can be run independently. To regenerate a single figure or table, invoke Snakemake with the specific target output file rather than the default all rule. The all rule currently targets only the data-sharing outputs (data_perm, data_nominal, mk_eqtl_tar); plotting and table rules are commented out pending manuscript finalisation.

Note

Data sharing archive

mk_eqtl_tar uses pigz for parallelised compression (--threads 8) and calls data_sharing_mk_eqtl_tar.sh to iterate over all cell types, collecting permutation and nominal eQTL files into the final archive. The FUSION TWAS weights archive (fusion_weights.tar.gz) is produced by the separate data_weights rule, which is also currently commented out of the all target.

Back to top