PSC-scDRS

PSC-scDRS is designed to link single-variant analysis scores with polygenic disease risk at single-cell RNA sequencing. It identifies cells with excess expression of disease-associated genes identified by Whole Exome Sequencing (WES).

In the Quick Start section, you can follow the instructions to install the requirements and run a sample. To learn more about the functions, refer to the flowcharts in the Flowcharts. Additionally, there, you can find detailed instructions on the data type and formats required as input for the PSC-scDRS function.

Table of contents

• Pipeline structure
• Quick Start
  • Prerequisites and Configuration
  • Downloads
  • Installing dependencies
  • Data
  • Running PSC-scDRS

Pipeline Structure

Quick Start

Prerequisites and Configuration

PSC-scDRS requires significant computational resources. Ensure your system meets the following minimum requirements.
System requirements for the sample dataset:
  - CPU: ≥ 16 cores
  - Memory: ≥ 32 GB RAM (scDRS may require up to 360 GB depending on dataset size),
  - Storage: ≥ 50 GB free disk space,
  - A stable high-speed internet connection for downloading approximately 22 GB of data.

** The pipeline with sample files was tested on a Dell XPS 15 9530 workstation running Ubuntu 22.04.5 LTS, equipped with 32 GB of RAM and a 13th-generation Intel® Core™ i9-13900H processor (20 cores).

Note: For large datasets, it is recommended to run the pipeline on a high-performance computing (HPC) system, as the scDRS method may require it, depending on the input files.

Download system requirements and repository

Download system requirements

sudo apt update
sudo apt install -y \
    git \
    build-essential \
    wget curl unzip \
    python3.12 python3.12-venv

Clone PSC-scDRS

All the codes and needed files for the sample file will be downloaded in this step.

cd ~
git clone https://github.com/ikmb/PSC-scDRS.git
cd PSC-scDRS
mkdir -p data
wget -O "data/HumanLiverHealthyscRNAseqData.zip" \
"https://github.com/seirana/PSC-scDRS/raw/main/data/HumanLiverHealthyscRNAseqData.zip"

Installing dependencies

PSC-scDRS needs some extra software to run:

step 1: Install scDRS

Pipeline will install it. If there is a problem, check their page.

step 2: Install the bcftools

Pipeline will install it. If there is a problem, check here.

step 3: Install MAGMA

Pipeline will install it. If there is a problem, check here.

This command installs Python libraries, scDRS, bcftools, and MAGMA.

cd ~
REPO_DIR="$(find . -maxdepth 5 -type f -name setup_dependencies.sh -path '*/PSC-scDRS/*' -print -quit | xargs -r dirname)"
echo "$REPO_DIR"
cd "$REPO_DIR"
bash ./setup_dependencies.sh

Data

Summary statistics for the GAISE single-marker test on PSC whole-exome sequencing data are available in the sampleWES.zip file.
The single-cell RNA sequencing data from the healthy human liver in the study by Andrews, T.S. et al. (PMID: 38199298) is provided as a sample dataset after applying the required modifications using the scDRS (PMID: 36050550) method, in the HumanLiverHealthyscRNAseqData.h5ad file.

Running PSC-scDRS

The pipeline will run the code smoothly.

cd ~
REPO_DIR="$(find . -maxdepth 5 -type f -name setup_dependencies.sh -path '*/PSC-scDRS/*' -print -quit | xargs -r dirname)"
echo "$REPO_DIR"
cd "$REPO_DIR"
bash ./PSC_scDRS_run.sh

** The final results are stored as ./PSC-scDRS/output/PSC cell association with Liver.csv