This repository contains the R code used for the analysis presented in the paper: "scInTime: A Computational Method Leveraging Single-Cell Trajectory and Gene Regulatory Networks to Identify Master Regulators of Cellular Differentiation" published in Genes 2022 (Read the paper). The study introduces a novel unsupervised machine learning framework that integrates single-cell RNA sequencing data with trajectory inference and gene regulatory networks to identify key regulators of cellular differentiation.
code/:CardiomyocyteDataAnalyse.r: Analysis of cardiomyocyte data.HNSCC6DataAnalyse.r: Analysis of head and neck squamous cell carcinoma data.ZebrafishDataAnalyse.r: Analysis of zebrafish data.
generate_networks/:- Contains scripts to generate gene regulatory networks at different pseudotime points.
data/:- Contains all data files used in the analyses, organized by study.
The analysis relies on several R packages which are crucial for the processing and analysis of the single-cell RNA-seq data:
Seurat: For preprocessing, normalization, and visualization of single-cell data.monocle3: Used for trajectory inference in single-cell data.scTenifoldNet: A package developed to build gene regulatory networks from single-cell data.- Other visualization and analysis packages:
ggplot2,ComplexHeatmap,circlize.
The analysis can be broken down into the following steps:
- Data Preprocessing: Raw data is processed using
Seuratfor quality control and normalization. - Trajectory Inference:
Monocle3is used to infer cellular trajectories and pseudotime. - Network Generation: Using
scTenifoldNet, gene regulatory networks are constructed for cells at different pseudotime stages. - Differential Expression Analysis: Identification of genes that show significant changes over pseudotime.
- Network Analysis: Analyze changes in gene regulatory networks across pseudotime to pinpoint key regulatory genes.
Please refer to each script for detailed implementation of these steps.