# Mfa Pipeline Orchestrator

Orchestrate the full metabolic flux analysis pipeline from model loading to phenotype...

- Repository: https://github.com/aiming-lab/AutoResearchClaw (14798 stars)
- Category: Engineering
- License: MIT
- Last updated: 2026-08-19
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add aiming-lab/AutoResearchClaw --skill mfa-pipeline-orchestrator -g -a claude-code -y`
Codex: `npx skills add aiming-lab/AutoResearchClaw --skill mfa-pipeline-orchestrator -g -a codex -y`
Cursor: `npx skills add aiming-lab/AutoResearchClaw --skill mfa-pipeline-orchestrator -g -a cursor -y`
Gemini CLI: `npx skills add aiming-lab/AutoResearchClaw --skill mfa-pipeline-orchestrator -g -a gemini-cli -y`

## Description

Orchestrate the full metabolic flux analysis pipeline from model loading to phenotype prediction and publication figures. Triggers when the user provides an organism name, BIGG model ID, or custom reaction list and wants end-to-end metabolic modelling run automatically.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# MFA Pipeline Orchestrator

## Overview

Coordinates all mfa-agent sub-agents in sequence, tracking progress via `progress/` markdown files so any failed step can be resumed independently.

**Full pipeline:**
```
Model source (BIGG ID / custom reactions)
  → [model-builder]     models/<Model>.json  +  validation report
  → [fba-runner]        simulations/fba_fluxes.csv  +  scan_summary.json
  → [flux-analyzer]     analysis/essentiality.csv  +  phase_plane.png
  → [metabolic-pheno-analyzer]  output/figures/*.pdf  +  yield table
```

## Workflow

### Step 0: Parse User Request

Extract and record in `progress/step0_inputs.md`:
- Model source (BIGG ID or custom)
- Organism and condition (aerobic/anaerobic, carbon source, concentration)
- Objective reaction (biomass or product)
- Gene knockouts to apply
- Analysis goals (essentiality, phase plane, yield optimisation, WT vs. mutant comparison)
- Target product (if yield analysis requested)

### Step 1: Invoke model-builder

Provide: model source, medium constraints, objective, knockouts.
Wait for `progress/step1_metabolic_model.md`.
Read: model file path, WT growth rate, model statistics.

### Step 2: Invoke fba-runner

Provide: model path, simulation types requested (FBA, pFBA, FVA, knockout screen), carbon source sweep if requested.
Wait for `progress/step2_fba_simulation.md`.
Read: flux CSV paths, essential gene count, secretion fluxes.

### Step 3: Invoke flux-analyzer

Provide: model path, FBA results, analysis goals (essentiality, phase plane, sampling), nutrient pair for phase plane.
Wait for `progress/step3_flux_analysis.md`.
Read: essential genes, phase plane optimum, engineering targets.

### Step 4: Invoke metabolic-pheno-analyzer

Provide: model path, all previous results, target product, publication requirements.
Wait for `progress/step4_metabolic_phenotype.md`.
Read: max theoretical yield, figure paths.

## Progress File Specification

The rest: https://github.com/aiming-lab/AutoResearchClaw/blob/HEAD/external/agents/Biology-Agent/skills/mfa-pipeline-orchestrator/SKILL.md