PARIST-Trial Workflow • Severe-Malaria AKI Project

End-to-end, reproducible code + text for the manuscript
“Paracetamol for Renal Recovery in Severe Malaria (PARIST Trial)”

---

📑 Table of contents

1. Project scope
   
2. Folder structure
   
3. Quick-start
   
4. Re-running the full pipeline
   
5. Continuous integration
   
6. Authorship & contributions
   
7. Data governance
   
8. License & citation
   
9. Contact

---

Project scope

This repository is the single source of truth for:

• De-identified analysis datasets
  
• R scripts and R Markdown notebooks
  
• Auto-generated figures & tables
  
• The manuscript in markdown/LaTeX (rendered to PDF & HTML)
  
• GitHub Actions CI to guarantee that analyses and rendering are repeatable

Why a public workflow?
Releasing code and de-identified data (+ exact package versions via renv) fulfils FAIR-data principles, improves transparency, and accelerates follow-up research.

---

Folder structure

```

parist-trial-workflow/ ├── analysis/ # R scripts & notebooks │ ├── 01_data-cleaning.R │ ├── 02_descriptive.Rmd │ ├── 03_RMST_analysis.R <- primary analysis │ └── utils/plot_theme.R ├── data/ │ ├── raw/ # 🔒 not in git (see .gitignore) │ └── derived/ # de-identified CSVs shared openly ├── manuscript/ │ ├── PARIST_manuscript.Rmd │ ├── references.bib │ └── journal-template.tex ├── figures/ # auto-generated PNG/PDF/SVG ├── docs/ # rendered manuscript (CI artifact / GitHub Pages) ├── archive/ # original Word files & approvals ├── .github/workflows/ci.yml ├── renv.lock # pinned R package versions ├── LICENSE | CITATION.cff └── README.md # you are here

````

---

Quick-start

Clone and recreate the R environment

```bash git clone https://github.com//parist-trial-workflow.git cd parist-trial-workflow

one-liner to restore exact packages

R -q -e "install.packages('renv'); renv::restore()" ````

Run the primary analysis

bash Rscript analysis/03_RMST_analysis.R

Render the manuscript locally

bash R -q -e "rmarkdown::render('manuscript/PARIST_manuscript.Rmd', output_dir = 'docs')" open docs/PARIST_manuscript.pdf

---

Re-running the full pipeline

| Stage | Script / tool | Output | | --------------------------------- | --------------------- | ------------------------------------------------------------------ | | 1 Import & clean | 01_data-cleaning.R | data/derived/parist_analysis_set.csv | | 2 Descriptive stats | 02_descriptive.Rmd | figures/Table1.png | | 3 Primary RMST + adjusted models | 03_RMST_analysis.R | figures/Fig2_RMST_curve.pdf, analysis/results/rmst_summary.csv |

The CI workflow executes exactly these steps on every push / PR to guarantee reproducibility.

---

Continuous integration

.github/workflows/ci.yml:

1. Checks out the repo.
2. Restores the renv cache.
3. Runs analysis scripts; fails if any exit non-zero.
4. Renders the manuscript to PDF.
5. Uploads artifacts & (optionally) deploys docs/ to GitHub Pages.

Status badge at the top of this README shows the latest build.

---

Authorship & contributions

We adopt the CRediT taxonomy. Example (edit as needed):

| Author | Contribution(s) | | ------ | -------------------------------------------------------- | | A.B. | Conceptualization, Methodology, Writing – original draft | | C.D. | Formal analysis, Software, Visualization | | E.F. | Investigation, Data curation | | G.H. | Supervision, Funding acquisition |

➡️ How to contribute:

1. Create a feature branch: git checkout -b feat/<slug>
2. Commit with Conventional Commit prefix (feat:, fix:, docs:…)
3. Open a Pull Request; CI must be green & 2 reviews obtained.
4. Squash-merge to main.

---

Data governance

| Data tier | Location | Policy | | --------------------- | --------------------------- | --------------------------------- | | Raw identifiable | Institution file-share only | Never pushed to GitHub | | Derived de-identified | data/derived/ | Commit OK; shared under CC-BY 4.0 | | Post-acceptance | Zenodo / Mendeley Data | DOI back-linked here |

Corresponding author stores the encryption key for raw ↔︎ derived mapping.

---

License & citation

• Code: MIT License – see LICENSE.
• Text & figures: Creative Commons CC-BY 4.0.

A ready-made CITATION.cff lets GitHub generate a citation snippet. When citing, please use the DOI badge at the top of this README once minted.

---

Contact

Corresponding author: Dr Paasi George – [georgepaasi8@gmail.com)

For issues or feature requests open a GitHub Issue or start a Discussion.

---

Happy analysing – and thanks for advancing open, reproducible clinical science!

```