Characterizing the extracellular matrix transcriptome of endometriosis

https://link.springer.com/article/10.1007/s43032-023-01359-w

Setup

Prerequisites

• Jupyter Notebook
• Python 3.7+
• R 4.2+

1. Install dependencies (R and Python packages)

The following R packages are installed automatically by the script, install_r_packages.r:

• affy
• sva
• readr
• dplyr
• Biobase
• BiocGenerics
• BiocParallel
• genefilter
• hgu133plus2cdf
• jsonlite
• org.Hs.eg.db
• stringi
• tibble
• limma
• yaml
• ggrepel
• devtools
• IRkernel
• clusterProfiler

Some of the listed R packages may require additional system dependencies.

If you have R set up to install packages system-wide (rather than to a personal user library), you can either run the install script as admin/superuser, or manually install the packages listed above (note that IRkernel is installed via devtools::install_github('IRkernel/IRkernel')).

Setup: Run the following commands at the command line:

git clone https://github.com/fogg-lab/characterizing-ecm-transcriptome-of-endometriosis.git
cd characterizing-ecm-transcriptome-of-endometriosis
pip install -r requirements.txt
Rscript install_r_packages.r

2. Prepare data for analysis

Run the Jupyter notebook, data_prep/prep.ipynb

Unsupervised analysis (hierarchical clustering)

Run the Jupyter notebook, analysis/clustering.ipynb

Condition stratification

Run the script:

cd analysis
python regression classifier.py

Compile condition stratification results and generate figures

Run the Jupyter notebook, analysis/get_classification_results.ipynb

Enrichment analysis

Run the Jupyter notebook, analysis/enrichment_analysis.ipynb

Differential expression analysis

Run the script, analysis/dgea.R

Usage

Rscript dgea.R <counts_filepath> <coldata_filepath> <config_filepath> [<filter_filepath>] <output_dir>

Example - Performing differential gene expression analysis with a filter list

In this example, we will run the dgea.R script with the following parameters:

• counts_filepath: The file all_phases_all_genes_counts.tsv contains count data.
• coldata_filepath: The file all_phases_coldata.tsv contains sample conditions, e.g. healthy/endometriosis.
• config_filepath: The YAML configuration file dgea_config.yaml is used.
• filter_filepath (optional argument): We are using the filter list core_matrisome_genes.json.
• output_dir: The results will be written to the dgea_output directory.

The command would be as follows:

Rscript analysis/dgea.R data/all/all_phases_all_genes_counts.tsv data/all/all_phases_coldata.tsv analysis/dgea_config.yaml analysis/core_matrisome_genes.json dgea_output

Command-line arguments (listed in positional order) for dgea.R

• -h or -help: Print usage information and exit.
• counts_filepath: Path to the file containing count data.
• coldata_filepath: Path to the file containing column data.
• config_filepath: Path to the YAML file containing configuration settings.
• filter_filepath: (Optional) Path to the JSON file containing gene filter list.
• output_dir: Directory where the output file will be written.