Mieru Splicing Project

This repository contains comprehensive RNA splicing analysis scripts and data for studying the effects of RNA-binding protein (RBP) knockouts on alternative splicing patterns in mouse embryonic stem cells. The project analyzes differential splicing events across 11 different RBP knockout lines compared to MIERU control cells.

Project Overview

The analysis investigates how specific RBP knockouts affect: - Alternative splicing patterns (5 event types: SE, A3SS, A5SS, MXE, RI) - Gene expression changes - Protein complex composition - Functional pathway enrichment

Key RBPs analyzed: Cd2bp2, Qk, Rbm24, Rpl22l1, Spen, Strap, Tra2b, Trim71, Ubr5, Wt1, Ybx1

Requirements

• Unix environment (WSL2 with Ubuntu or macOS recommended)
• conda package manager via miniforge
• ~50GB storage space for genome indices and intermediate files

Installation

Create and activate the conda environment with all required bioinformatics tools:

bash conda env create -f environment.yml conda activate mieru

[!IMPORTANT] To ensure full reproducibility of this project's analysis results, the following files are provided: - environment.yml: Complete conda environment (including build numbers) - environment-no-builds.yml: Cross-platform compatible version - conda-packages-list.txt: Detailed list of installed packages - R-session-info.txt: R session information - REPRODUCIBILITY.md: Step-by-step instructions for reproducibility
For details, see REPRODUCIBILITY.md.

Dataset Access

Raw sequencing data and processed results are publicly available:

• Fig2 data: Control samples and marker gene analysis

  • FASTQ files: GSE291522
  • Location: Fig2/data/fastq/

• Fig3-6 data: RBP knockout comparative analysis

  • FASTQ files: GSE291672
  • rMATS splicing results: GSE291672_rmats.zip
  • Location: Fig3-6/data/fastq/ and Fig3-6/data/rmats/original_output/

Pipeline Overview

The analysis pipeline follows a modular, standardized organization:

Fig2 Analysis (Control Characterization)

• Fluorescent Analysis (025-fluorescents/): Fluorescent protein expression quantification
• Marker Gene Analysis (035-marker_genes/): Validation of cell line markers

Fig3-6 Analysis (Main Pipeline)

1. Setup (00-setup/): System dependencies, directory creation, and genome data download
2. Preprocessing (01-preprocessing/): Quality control, trimming, genome alignment, read counting, and differential splicing with rMATS
3. Quality Control (02-quality-control/): Exon characteristics analysis and quality metrics
4. Event Analysis (04-event-analysis/): Alternative splicing event frequency and ΔPSI distribution analysis (Figure 3)
5. DEG Comparison (05-deg-comparison/): Integration with differential gene expression (Figure 4)
6. Complex Analysis (06-complex-analysis/): Protein complex enrichment using CORUM/ComplexTab databases (Figure 5)
7. Heatmap Analysis (07-heatmap-analysis/): GO term and pathway visualization (Figure 6)

Legacy Structure

• Original organization preserved in _past/ directories for backward compatibility
• Legacy scripts (e.g., 015-preprocess/, Fig3-event-frequency/) remain functional

Usage

1. Download raw data from GEO repositories
2. Activate conda environment: conda activate mieru
3. Run preprocessing scripts in numerical order within each analysis directory
4. Scripts must be executed from their respective directories due to relative paths