Single-Cell Analysis Pipeline for ONT Long Reads

This repository houses a comprehensive Snakemake workflow designed for the analysis of single-cell sequencing data from Oxford Nanopore Technologies (ONT). It seamlessly integrates data processing using ONT's wf-single-cell Nextflow pipeline, downstream analysis with standard R packages for single-cell analysis, and utilizes SQANTI3 for detailed isoform classification.

Pipeline Overview

The pipeline processes single-cell sequencing data through the following major stages:

1. ONT wf-single-cell: Utilizes the ONT Nextflow pipeline to process raw reads into matrix files.
2. Single-Cell Data Analysis: Implements R packages like Seurat and SingleR for cell clustering and typing.
3. Isoform Classification and Quantification: Uses SQANTI3 to classify cell-specific isoforms, followed by merging the quantification data back onto the reads.
4. Visualization and Reporting: Generates plots and detailed reports for both cluster analysis and isoform classification.

Prerequisites

Software Requirements

• Snakemake: Orchestrates and manages the workflow execution.
• Nextflow: Required for running the ONT-specific workflows.
• Singularity: Ensures reproducibility by containerizing the software environments.
• R: A statistical computing environment where several analysis scripts are executed. Ensure the following libraries are installed:
  • Seurat
  • SingleR
  • clusterProfiler
• Python: Needed particularly for SQANTI3 and data merging scripts.
• Conda: Recommended for managing Python and R environments seamlessly.

System Requirements

• Access to a computational environment with a substantial amount of memory and CPU resources.
• SLURM or another job scheduler for managing job submissions.

Installation

Clone this repository to your local machine or computational cluster:

git clone https://github.com/[YourUsername]/single-cell-ont-analysis.git
cd single-cell-ont-analysis

Ensure all dependencies are installed by setting up environments through Conda:

conda env create -f environment.yml

Configuration

Modify the config.yaml file to reflect your project-specific settings including paths, reference genomes, and tool parameters. Sample configurations and environment setup details are also provided for clarity.

Running the Pipeline

To execute the pipeline, use the following command:

sbatch submit.sh

Outputs

The pipeline will generate the following outputs:

• Gene expression matrices in .mtx.gz format.
• R objects for downstream analysis (.rds).
• Classified isoforms and comprehensive reports in HTML format from SQANTI3.
• Final merged classification data, linking quantifications to read data.

Detailed Scripts Description

Scripts located in scripts are critical for various stages of the pipeline, including data merging and additional analyses not handled directly by Snakemake.

For any questions, contact

CCRSF_IFX@nih.gov

Acknowledgments

• Thanks to EPI2ME Labs for providing the wf-single-cell Nextflow pipeline.
• Gratitude to the developers of Seurat, SingleR, and SQANTI3 for their open-source software which this pipeline depends on.