This repository contains a complete pipeline for analyzing single-cell RNA-seq (scRNA-seq) data in broccoli (Brassica oleracea) using the Seurat package for clustering and Monocle 3 for pseudotime/trajectory analysis.
The aim is to investigate gene expression heterogeneity and senescence progression in broccoli at the single-cell level, particularly focusing on preharvest and postharvest conditions. This pipeline enables:
- Quality control and filtering
- Normalization and dimensionality reduction
- Cell clustering and UMAP visualization
- Marker gene identification
- Pseudotime trajectory analysis
The input dataset is generated using 10X Genomics single-nucleus RNA-seq (snRNA-seq) on broccoli inflorescence tissue. It includes the following files typically found in the filtered_feature_bc_matrix directory:
matrix.mtx
barcodes.tsv.gz
features.tsv.gz
Install the required R packages:
install.packages("Matrix")
install.packages("Seurat")
if (!requireNamespace("BiocManager", quietly = TRUE)) install.packages("BiocManager")
BiocManager::install("monocle3")Clone this repository:
git clone https://github.com/your-username/broccoli-scRNA-seq.git
cd broccoli-scRNA-seqOpen and run the pipeline in R:
source("broccoli_seurat_pipeline.R")Make sure to update the setwd() path in the script to where your data files are located.
cluster_markers.csv: Top marker genes per clusterpseudotime_values.csv: Pseudotime values for each cell
- Violin plots of QC metrics (
nFeature_RNA,nCount_RNA) - PCA and Elbow plots
- UMAP with clusters
- Top variable features plot
- Pseudotime trajectory plot
This pipeline is designed to identify senescence-associated genes (SAGs) and understand cell type-specific responses during postharvest senescence. By integrating pseudotime analysis, we aim to reconstruct the senescence trajectory and reveal regulatory mechanisms underlying broccoli quality decline after harvest.