AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

RELSA Severity Assessment & Humane Endpoint Forecasting

Combines lab-animal welfare readouts into a single RELSA severity score and forecasts humane endpoints with ARIMA plus KDE-derived severity zones.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

  • Merges multiple welfare readouts (weight loss, body temperature, clinical/nesting scores, biomarkers, activity, heart rate, burrowing) into one RELSA score per animal per time point, relative to a reference set of known burden (0 = baseline, 1 = the reference set's maximum deviation).
  • scripts/relsa_score.py builds, echoes and serialises the reference model to JSON so later cohorts stay on the same scale.
  • scripts/forecast_relsa.py fits an auto-selected ARIMA (Hyndman–Khandakar AICc) to each animal's trajectory, predicts the next score with a 95% interval, and reports RMSE, PICP and MPIW together.
  • scripts/kde_thresholds.py finds density minima by kernel density estimation to propose attention/danger zones, with a mandatory bandwidth sensitivity sweep.
  • Makes the four result-determining decisions explicit (directionality of "turned" variables, reference set choice, mapping zero-baseline ordinal scores, stable variable composition) and supplies a reporting checklist plus hard boundaries (not a regulatory severity grading, not a validated predictor of death).

Who it's for

  • Researchers writing severity-assessment, 3Rs/refinement or EU Directive 2010/63/EU sections
  • Animal welfare bodies / IACUC members needing quantitative support for humane endpoint decisions
  • Data analysts handling telemetry, body-weight and clinical-score time series in preclinical studies

Example uses

  1. Run relsa_score.py on a cohort CSV to get per-animal per-day RELSA scores with per-variable weights, saving the endpoint-reaching group as reference.json for reuse.
  2. Run forecast_relsa.py --animals M01,M02 --endpoints M01=5 to train up to the step before the endpoint, predict the endpoint score, and evaluate with RMSE/PICP/MPIW; switch to --mode rolling for daily one-step-ahead monitoring.
  3. Run kde_thresholds.py --n-thresholds 2 to derive attention/danger cutoffs from density minima, then report the bandwidth and its sensitivity sweep in the methods.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/relsa-severity-assessment/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the skills/relsa-severity-assessment folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/relsa-severity-assessment/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/relsa-severity-assessment ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and confirm Python 3.10+: python --version
  2. Install dependencies: uv pip install "numpy>=1.26" "pandas>=2.0" "scipy>=1.11" "statsmodels>=0.14" matplotlib (or pip install ... if you don't use uv)
  3. Clone the repository: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  4. Copy the skill into Claude's skills folder: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/relsa-severity-assessment ~/.claude/skills/
  5. Restart Claude Code and ask something like "compute a RELSA severity score from these weight, temperature and clinical-score readings".
  6. For a first smoke test, run the Step 1 command from the docs against the bundled assets/example_cohort.csv.
View source on GitHub ↗License: MIT