Enrichment analysis of SysMalVac gene expression measurements

This repository contains all the scripts used to produce the enrichment analysis of Moncunill et al. (2019) manuscript.

• Input data, intermediate data and a raw version of the results can be downloaded from Figshare.
• Final enrichment results can be downloaded from Figshare, too: sysmalvac_deliverable.zip file.
• An even more succint version of the data is provided in the manuscript as Supplementary Data in the manuscript.

For specific questions about the scripts, please send an email to miquel.duran@irbbarcelona.org. For general questions about the SysMalVac study, please contact gemma.moncunill@isglobal.org or carlota.dobano@isglobal.org.

Specifications

All the scripts necessary to run the analysis are provided within this repository. The procedures involve many enrichment analyses. Most processes were run with a local SGE cluster at IRB Barcelona. The main script to set up jobs is ./setupArrayJob.py. We provide a Singularity image containing the necessary dependencies. The folder contains Python 2.7 (*.py), IPython Notebooks (*.ipynb) and R-package (*.R) scripts. Scripts are prepared to run with Linux systems.

This repository is provided to ensure data transparency. The outcome of the scripts is conveniently provided as compressed files. To complete the full repository in your home directories, please download the corresponding files from Figshare and uncompress them inside the cloned repository folder:

• data.tar.gz
• sysmalvac_results.tar.gz
• sysmalvac_results_wgcna.tar.gz

Guide through the scripts

1. The Parsing.ipynb notebook was used to provide some gene mappings, given the differential gene expression analysis. Input data for the scripts below is provided as ./data.
2. SGE scripts are denoted by a _ before the script name. At IRB, queueing systems are called all.q and fast.q. This should be edited at the end of the scripts if another cluster is used.
   • _enrichment.py handles the GSEA analysis against modules and gene sets.
   • _enrichr.py handles the EnrichR analysis.
   • _enrichment_wgcna.py handles the GSEA analysis based on co-expression (WGCNA) analysis (see below).
3. Locally-run R scripts. These are light-computation scripts that do not require SGE computation.
   • wgcna.R performs WGCNA analysis based on the DCGL package.
   • camera_and_tmod.R performs Camera and TMod analyses to complement GSEA.
4. Running the scripts above produces a complex folder structure.
   • sysmalvac_results folders contain the GSEA, Tmod, Camera and EnrichR analysis of differential expression (conventional enrichment).
   • sysmalvac_results_wgcna folders contain the GSEA results of the differential co-expression analysis.
5. Summarizing results.
   • summarize.ipynb produces a summary table of the conventional enrichment results.
   • summarize-wgcna.ipynb produces a summary table of the WGCNA enrichment results.

For convenience, we manually assembled a deliverable (sysmalvac_deliverable.zip) containing the relevant input data, conventional enrichment results (sysmalvac_diff_expr) and WGCNA enrichment results (sysmalvac_wgcna).