SCERC2 data analysis

Code required to reproduce [Cheng and Morris et al.,eClinicalMedicine 2025].

folder structure

• DEQA_templates = downloaded version of the data extraction and quality assessment table that our team members see on Covidence.
• Code = any code, including bash, python, R scripts and notebooks.
• data = The downloaded .csvs from Covidence
  • data extraction table when downloaded requires a suffix change to DE.csv to ensure uniqueness to find
  • please AVOID changing the files manually, and only read into and write from scripts to maximise reproducibility. When reading in to R, be reminded to set read.csv(check.names = FALSE) to avoid characters being altered.
  • /lineage is data used for lineage imputation.
• output
  • contains figures, intermediate and final tables.
  • Tables that are intrinsic to the Data Extraction template are in /output/processed_tables, they include:
    • Baseline characteristics (Age and sex distribution of the cancer and non-cancer cohort respectively)
    • Co-morbidities
    • outcome tables which includes
      • Primary Cancer Site
      • Metastasis status
      • Treatment type
      • COVID variants
      • COVID vaccination status
  • the exception being /data/review_241469_20240110223959_DE_knownDx_peerReviewed.csv which is generated by data/0.1_Cleaning.Rmd as it is just a cleaning script.
• team_members_automated.xlsx