anu-bioinfo / pancanmet_analysis

Software and analysis of metabolism studies across multiple cancer types

Pan-Can Metabolism Analysis

Input data and informatic pipeline, analysis, visualization R code for the analysis of metabolomics data from over 900 tissue samples spanning 7 cancer types.

Code Structure

• analysis: R code for analyses performed in the project. Scripts prefixed by Run* are primary scripts used in the project analysis. Some of the key scripts are described below:
  • RunVisualizeClinical.R: Visualize clinical data
  • RunSummary: Generate a summary of the merged metabolomics data
  • RunKeggCoverage: Calculate the coverage of KEGG represented by the project dataset
  • RunPathwayCommonsCoverage: Calculate the coverage of Pathway Commons represented by the project dataset
  • RunPharmaCoverage: Determine enzymes making use of metabolites (as substrates or products) in the study are targetable by drugs
• data: Primary data that was processed as part of the project
  • merged_metabolomics: Merged datasets after informatic standardization (e.g. name mapping across studies) pipeline, including metabolic profiling values, clinical features, and paired tumor-normal fold changes
  • studies: Primary data collected from studies (often supplementary tables) both metabolomic profiling and clinical variable data
  • pharmacology_drugbank: CHEBI IDs from Pathway Commons (pathwaycommons.org) dataset categorized using DrugBank drug categories
• import: Scripts for importing data (e.g. KEGG pathway data, metabolite ID mapping)
  • RunMergeMetabolomics: Merges the individual metabolomics files,
  • RunManualImport: Imports metabolomics data from a few studies that do not have normalized data
  • RunMap2Kegg: Maps our metabolite data to KEGG IDs
• results: Results from analysis
• shinyapp: Data used for the R Shiny web application