Assessing the Impact of Copy Number Variation on Gene Regulatory Network Inference in scRNA-seq

Description

In this project, I am to understand how copy number variations (CNVs) influence gene regulatory network (GRN) inference in single-cell RNA sequencing (scRNA-seq). There are two outcomes as of now: - Quantitative analysis of CNVs' impact on scRNA GRN inference, highlighting the limitations and robustness of current scRNA GRN inference methods under genomic instability - Identification of biological pathways and GO terms that are especially affected by CNVs in scRNA datasets, providing insights into potential biases or current scGRN methods.

Installation

To ensure reproducibility and ease of setup, YAML files have been provided to configure the necessary dependencies. Simply use your preferred environment manager to create the required setup:

conda env create -f environment.yaml conda activate <env_name>

Pipeline

The analysis consists of three main components:

1. CNV calling - 2 methods are used to call the CNVs from the scRNA-seq data (the output of inferCNV is used for downstream visualization)
   • copyKAT
   • inferCNV
2. GRN inference – GRNs are created based on the scRNA-seq data
   • SCENIC
3. Integration and visualization

Usage

To run the integrated pipeline for copy number variation (CNV) and gene regulatory network (GRN) analysis on scRNA-seq data, follow these steps:

Prerequisites

Ensure that you have access to the necessary computational resources and that the required software dependencies are installed. You will need: - Conda environment: Install and activate the appropriate Conda environments (e.g., copyKAT, r_env2, pyscenic). - Input Data: The script assumes that the required input data is already available in the specified directory paths.

Running the Pipeline

1. Clone the repository (if applicable):
   ``` git clone https://github.com/your-repo-url.git cd your-repo-directory
2. Set up the conda environments
   The required environments are listed in the Conda YAML files provided (e.g., environment.yaml). Use the following command to set them up: ``` conda env create -f environment.yaml conda activate
3. Prepare the sample and dataset configuration Modify the variables DATASETID and SAMPLEID in the script to match your data. For example: ``` DATASETID="ccRCCGBM"
   SAMPLEID="C3L-00004-T1CPT0001540013"
4. Submit the job to SLURM This script is intended to be run on a SLURM-based cluster. You can submit the job using the following command: sbatch sc_analysis_pipeline.sh

5. Monitor progress
   Monitor the progress of the job by checking the output and error logs:

   • Standard output: /path/to/your/dir/utilities/logs/integrated_pipeline.out
   • Error output: /path/to/your/dir/utilities/logs/integrated_pipeline.err

6. Output directories The results from each pipeline will be saved in the corresponding output directories:

   • copyKAT results: ${BASE_DIR}/scCNV/copyKAT/${DATASET_ID}/${SAMPLE_ID}
   • inferCNV results: ${BASE_DIR}/scCNV/inferCNV/${DATASET_ID}/${SAMPLE_ID}
   • pySCENIC results: ${BASE_DIR}/scGRNi/RNA/SCENIC/${DATASET_ID}/${SAMPLE_ID}

Customization

You can modify the following parameters within the script to adjust the pipeline for different datasets or analysis configurations: - CELLTYPES: Specifies the cell types to be used in pySCENIC analysis (e.g., "None", "Tumor", "Non-Tumor"). - PRUNEFLAGS: Specifies pruning options for pySCENIC (e.g., "None"). - The default is that pruning is turned on, with no special folders being made. - Once pruning is specifically turned on or off (e.g. "True" or "False"), specific folders will be made where you can find the output ("pruned" or "unpruned")