RUNX3 orchestrates a context specific transcriptional switch with opposite outcomes in gastric cancer metastasis
Description
This repository contains the bioinformatics analysis code used in the study “RUNX3 orchestrates a context-specific transcriptional switch with opposite outcomes in gastric cancer metastasis.” All scripts and pipelines are organized into four top-level folders: • RNAseq_Code: RNA-seq downstream analysis code for metastatic gastric cancer cell lines HGC27 (p53-deficient/LOF), LMSU (TP53 R175H GOF), and NUGC4 (wild-type p53; ARF-defective), including analyses of TGFβ-stimulated conditions and inhibitor-perturbation datasets. The workflows cover differential expression (DESeq2-based), gene set/pathway analyses (e.g., GSEA/GO), EMT/apoptosis-related scoring, and figure-ready visualizations. • ChIPseq_Code: ChIP-seq analysis code for HGC27 cells, including processing and downstream analyses of RUNX3 and chromatin-mark ChIP-seq datasets (e.g., peak calling, annotation, metagene/heatmap profiling, and motif/enrichment analyses) aligned to the human reference genome (GRCh38/hg38). • TMA_Code: Code for tissue microarray (TMA) quantification and statistical analyses used in the manuscript, including nuclear scoring metrics (e.g., H-score/positivity) and figure generation for the two gastric cancer TMAs (ST8013b and ST810d). • 74_liver_metastases_Code: Code used to analyze an independent institutional cohort of 74 resected gastric cancer liver metastases, including RUNX3 IHC intensity/positivity assessments and stratified analyses by histological differentiation grade. This deposit provides analysis code (scripts/pipelines) used to generate the processed results and figure-level summaries reported in the manuscript. Raw sequencing data are described in the manuscript. For questions regarding the analysis code, please contact the corresponding author (Yoshiaki Ito; yoshi_ito@nus.edu.sg).
Files
Steps to reproduce
• RNA-seq: raw FASTQ files were quality-checked, aligned/quantified against GRCh38/hg38, and summarized to gene-level counts. Differential expression was performed in R (DESeq2), followed by pathway/gene set analyses (GSEA/GO; clusterProfiler) and figure generation (ggplot2 and related packages). • ChIP-seq: HGC27 ChIP-seq reads were processed and aligned to GRCh38/hg38, peaks were called, and peaks were annotated and visualized (metagene/heatmaps and locus-level tracks). • TMA and 74-cohort IHC: nuclear staining was quantified using predefined scoring criteria (e.g., intensity/positivity and/or H-score). Statistical analyses and plots were generated in R. • Software: R (version 4.3.3), and key packages/tools used in each folder are documented within scripts.
Institutions
- National University of Singapore