Metabolomics Statistics

---
name: metabolomics-statistics
description: >-
  Statistical analysis for metabolomics — PCA, PLS-DA, clustering, and univariate tests.
version: 0.1.0
author: OmicsClaw
license: MIT
tags: [metabolomics, statistics, PCA, clustering]
metadata:
  omicsclaw:
    domain: metabolomics
    emoji: "📈"
    trigger_keywords: [metabolomics statistics, multivariate, PCA, clustering]
    allowed_extra_flags: []
    legacy_aliases: [met-stat]
    saves_h5ad: false
---

# 📈 Metabolomics Statistical Analysis

Statistical analysis module for metabolomics data. PCA, PLS-DA, hierarchical clustering, and univariate tests.

## CLI Reference

```bash
python omicsclaw.py run met-stat --demo
```

## Why This Exists

- **Without it**: Metabolomic variance is inherently high-dimensional and non-trivial to dissect
- **With it**: Advanced clustering algorithms and projections distill variance into biologically valid groups
- **Why OmicsClaw**: Wraps complex R/Bioconductor modules into a clear Python execution syntax

## Workflow

1. **Calculate**: Compute distance matrices (Euclidean, Pearson).
2. **Execute**: Project high-dimensional structures via PCA/t-SNE/UMAP.
3. **Assess**: Execute hierarchical clustering mapping samples to metabolic profiles.
4. **Generate**: Output coordinate projections.
5. **Report**: Synthesize scree plots, scatter projections, and heatmaps.

## Example Queries

- "Run PCA on my normalized metabolomics data"
- "Perform hierarchical clustering with Ward's method"

## Output Structure

```
output_directory/
├── report.md
├── result.json
├── statistics.csv
├── figures/
│   ├── pca_projection.png
│   └── sample_heatmap.png
├── tables/
│   └── principal_components.csv
└── reproducibility/
    ├── commands.sh
    ├── requirements.txt
    └── checksums.sha256
```

## Safety

- **Local-first**: Strict offline processing without external upload.
- **Disclaimer**: Requires OmicsClaw reporting structures and disclaimers.
- **Audit trail**: Hyperparameters and operational flow states are logged fully.

## Integration with Orchestrator

**Trigger conditions**:
- Automatically invoked dynamically based on tool metadata and user intent matching.

**Chaining partners**:
- `met-normalize` — Upstream data scaling
- `met-diff` — Parallel structural differential assessment

## Citations

- [MetaboAnalystR](https://doi.org/10.1093/bioinformatics/bty528)
- [ropls](https://doi.org/10.1021/acs.jproteome.5b00354)