HSC scRNA-ATAC-seq Analysis

Overview

This repository contains the analysis and results of a single-cell multi-omics study combining scRNA-seq and ATAC-seq data to investigate hematopoietic stem cells (HSCs) across different age groups. The analysis focuses on the transcriptional and chromatin accessibility landscapes in young and aged HSCs, with a particular emphasis on age-related changes in chromatin accessibility and stress response pathways.

Project Structure

single_cell/
├── README.md                 # Main project documentation
├── scripts/                  # Analysis scripts
├── data/                     # Raw and processed data
├── plots/                    # Generated visualizations
├── output/                   # Analysis outputs
├── results/                  # Processed results
├── reports/                  # Generated reports
└── docs/                     # Documentation

For detailed project structure, see PROJECT_STRUCTURE.md.

Quick Start

1. Clone this repository
2. Install required R packages:

   install.packages('BiocManager')
   BiocManager::install(c("Signac", "Seurat", "EnsDb.Mmusculus.v79"))

3. Download data from GEO (GSE190424)
4. Run the analysis pipeline:

   rmarkdown::render("scripts/HSC_scRNA-ATAC_seq.Rmd")

Data Source

The data used for this analysis is publicly available on GEO under the accession GSE190424. The study provides high-resolution single-cell datasets, allowing detailed insights into HSC biology.

Original Research

This work is a reproduction of the research presented in the paper:
"Epigenetic traits inscribed in chromatin accessibility in aged hematopoietic stem cells" by Itokawa et al., published in Nature Communications (2022).
DOI: 10.1038/s41467-022-30374-9 | PMID: 35577813

Analysis Overview

Objectives

1. Compare gene expression profiles between young and aged HSCs
2. Identify chromatin accessibility differences linked to aging
3. Integrate scRNA-seq and ATAC-seq datasets to understand age-related changes in gene regulation and stress responses

Key Findings

• Chromatin Accessibility Changes: Aged HSCs exhibit alterations in chromatin accessibility, particularly in regions enriched for STAT, ATF, and CNC family transcription factor motifs
• Stress Response Pathways: Open differentially accessible regions (open DARs) in aged HSCs are linked to augmented transcriptional responses under stress conditions
• Enhancer States: Most open DARs comprise active, primed, and inactive enhancers, suggesting a role in regulating stress-induced gene expression

Analysis Pipeline

1. Data Preprocessing

• scRNA-seq Data:
  • Quality control filtering
  • Normalization
  • Feature selection
• ATAC-seq Data:
  • Peak calling
  • Fragment file processing
  • Quality metrics calculation

2. Quality Control

• Cell filtering based on:
  • Peak region fragments (3000-20000)
  • Percentage of reads in peaks (>15%)
  • Blacklist ratio (<0.05)
  • Nucleosome signal (<4)
  • TSS enrichment (>2)

3. Analysis Steps

1. Dimensionality Reduction

   • PCA
   • UMAP
   • t-SNE

2. Clustering

   • Graph-based clustering
   • Cluster marker identification

3. Differential Analysis

   • Gene expression differences
   • Chromatin accessibility changes
   • Motif enrichment analysis

4. Integration

   • RNA-ATAC data integration
   • Correlation analysis
   • Regulatory network inference

Results

1. UMAP Clustering Analysis

UMAP visualization showing distinct clustering patterns between young and aged HSCs. The analysis reveals clear separation between different cell populations, indicating distinct chromatin states and transcriptional profiles.

2. Gene Expression Analysis

Feature plot showing expression patterns of key genes across different cell clusters. This visualization helps identify genes that are differentially expressed between young and aged HSCs.

3. Expression Level Distribution

Violin plot showing the distribution of expression levels for key genes. This analysis reveals how gene expression varies between young and aged HSC populations.

4. Chromatin Accessibility Profiles

Coverage plot showing chromatin accessibility profiles around differentially accessible regions. This visualization demonstrates how chromatin accessibility changes with age in specific genomic regions.

5. Transcriptional Profiles

• Aged HSCs exhibited altered expression of genes associated with:
  • Stemness (e.g., GATA2, RUNX1)
  • Differentiation
  • Stress response

6. Chromatin Accessibility

• Aged HSCs displayed:
  • Reduced accessibility in promoter and enhancer regions
  • Open DARs enriched for specific transcription factor motifs
  • Altered regulatory landscape

7. Integrated Analysis

• Strong correlation between chromatin accessibility and gene expression
• Enhanced transcriptional responses under stress conditions
• Regulatory network changes in aged HSCs

Technical Details

Prerequisites

• R (version 4.0 or higher)
• Required R packages:
  • Seurat (for scRNA-seq analysis)
  • Signac (for ATAC-seq analysis)
  • EnsDb.Mmusculus.v79 (for genome annotation)
  • Additional packages: tidyverse, ggplot2

Data Requirements

• ATAC-seq peak count matrix (.h5)
• Cell metadata (.csv)
• Fragment files (.tsv.gz)
• Genome annotation (mm10)

Output Files

• Processed data objects
• Quality control plots
• Differential analysis results
• Integration analysis results
• Visualization plots

Troubleshooting

Common Issues

1. Memory Issues

   • Solution: Increase R memory limit
   • Command: memory.limit(size = 16000)

2. Package Installation

   • Solution: Install BiocManager first
   • Command: install.packages('BiocManager')

3. Data Loading

   • Solution: Check file paths and formats
   • Ensure all required files are present

Performance Tips

• Use parallel processing for large datasets
• Implement chunked processing for memory efficiency
• Save intermediate results

Citations

If you use this repository or the reproduced analysis in your work, please cite:

Itokawa N, Oshima M, Koide S, Takayama N et al. Epigenetic traits inscribed in chromatin accessibility in aged hematopoietic stem cells. Nat Commun 2022 May 16;13(1):2691. PMID: 35577813

License

This project is licensed under the MIT License. See docs/LICENSE for details.

Contact

For questions or issues, please open an issue in this repository.

Acknowledgments

• Original paper authors for providing the dataset
• Seurat and Signac development teams
• R/Bioconductor community