Supplementary code archive manifest

00_run_all.R
  Convenience runner for the four analysis and figure scripts.
01_primary_and_sensitivity_analysis.R
  Primary 24-hour landmark coefficient analysis, descriptive and sensitivity
  analyses, and exploratory-score bootstrap.
02_score_selection_adjustment.R
  Exploratory score construction and selection-adjusted parametric bootstrap.
03_validation_before_imputation_internal_validation.R
  Ordinary bootstrap optimism correction with multiple imputation nested
  within resampling. In each iteration, separate MICE runs complete the
  incomplete bootstrap sample and the original incomplete cohort. For each
  model and performance metric, bootstrap-sample and original-cohort
  performance are averaged separately across five imputation indices, and
  resample-specific optimism is calculated as the difference between those
  two averages. The script also derives and plots Supplementary Figure S3 from
  the bootstrap output.
04_regenerate_updated_figure2.R
  Regenerates Figure 2 and panel-level source data from the corrected internal-
  validation results.
README.txt
  Data-access, software, run-order, and runtime instructions.
analysis_data_template.csv
  Empty standardized template for the main de-identified dataset.
timing_audit_template.csv
  Empty standardized template for the detailed timing audit.
analysis_data_dictionary.csv
  Variable definitions, coding, and units.
expected_key_results.csv
  Principal aggregate values used to verify a reproduction run.
sessionInfo_authors.txt
  R and package versions used for the previous verified author run.
sessionInfo_validation_before_imputation.txt
  R and package versions used for the corrected validation-before-imputation
  analysis reported in this revision.
aggregate_validation_results/
  Corrected aggregate performance, between-model differences, and exact
  Figure 2/Figure S3 plotting coordinates from the verified author run.

No patient-level data or identifiers are included in this archive.
