agentleFS
dailycafidailycafi/metabolism-skills/skills/systems-biology/metabolic-reconstruction/SKILL.md

bio-systems-biology-metabolic-reconstruction

Skill7 starsChanged 8 months agoMITView on GitHub

Build genome-scale metabolic models (GEMs) from genome sequences using CarveMe and gapseq for automated reconstruction. Use when: user wants to build a metabolic model from a genome, create a GEM for a new organism, reconstruct metabolism from FASTA/GenBank, or generate a draft SBML model. Triggers: build metabolic model, genome-scale model, CarveMe, gapseq, metabolic reconstruction, GEM, draft model, SBML generation, organism-specific model, de novo metabolic model, reconstruct metabolism.

  • Installs packages

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before using code patterns, verify installed versions match.

    line 16
  2. Distinctive:

    If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API …

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Version Compatibility

Reference examples tested with: COBRApy 0.29+, NCBI BLAST+ 2.15+

Before using code patterns, verify installed versions match. If versions differ:

  • Python: pip show <package> then help(module.function) to check signatures
  • CLI: <tool> --version then <tool> --help to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Metabolic Reconstruction

"Build a metabolic model for my organism from its genome" → Generate a genome-scale metabolic model by mapping protein sequences to a universal reaction database, then gap-filling to ensure growth capability on specified media.

  • CLI: carve (CarveMe) for automated reconstruction from protein FASTA
  • CLI: gapseq find + gapseq draft + gapseq fill for pathway-based reconstruction

CarveMe (Recommended)

# Install CarveMe
pip install carveme

# Basic reconstruction from protein FASTA
carve genome.faa -o model.xml

# Specify output format
carve genome.faa -o model.xml --format sbml
carve genome.faa -o model.json --format json

# Gap-fill for specific media
carve genome.faa -o model.xml --gapfill M9

# Available media: M9, LB, M9[glc], M9[glyc], etc.

CarveMe Options

# Use diamond instead of blastp (faster)
carve genome.faa -o model.xml --diamond

# Specify organism type
carve genome.faa -o model.xml --grampos  # Gram-positive
carve genome.faa -o model.xml --gramneg  # Gram-negative (default)

# Initialize from template model
carve genome.faa -o model.xml --init M9

# Verbose output for debugging
carve genome.faa -o model.xml -v

gapseq (Alternative)

# Install gapseq
git clone https://github.com/jotech/gapseq
cd gapseq
./gapseq check  # Check dependencies

# Full reconstruction workflow
./gapseq find -p all genome.fasta  # Find metabolic pathways
./gapseq find -t all genome.fasta  # Find transporters
./gapseq draft -r genome-all-Reactions.tbl \
               -t genome-Transporters.tbl \
               -p genome-all-Pathways.tbl \
               -c genome.fasta
./gapseq fill -m genome-draft.RDS -c genome.fasta -n M9

Python API for CarveMe

import subprocess

def reconstruct_model(fasta_path, output_path, media='M9', grampos=False):
    '''Run CarveMe reconstruction

    Args:
        fasta_path: Path to protein FASTA file
        output_path: Output model file path (.xml or .json)
        media: Gap-filling media (M9, LB, etc.)
        grampos: True for Gram-positive organisms

    Model size expectations:
    - Bacteria: 1000-2500 reactions typical
    - Fungi: 1500-3000 reactions
    - Archaea: 800-1500 reactions
    '''
    cmd = ['carve', fasta_path, '-o', output_path, '--gapfill', media]

    if grampos:
        cmd.append('--grampos')

    subprocess.run(cmd, check=True)
    return output_path

Load and Inspect Draft Model

import cobra

model = cobra.io.read_sbml_model('model.xml')

print(f'Reactions: {len(model.reactions)}')
print(f'Metabolites: {len(model.metabolites)}')
print(f'Genes: {len(model.genes)}')

# Check if model can grow
solution = model.optimize()
print(f'Growth rate: {solution.objective_value:.4f}')

# List exchange reactions (available nutrients)
for rxn in model.exchanges[:10]:
    print(f'{rxn.id}: {rxn.reaction}')

Quality Metrics

def assess_model_quality(model):
    '''Basic quality assessment for draft model

    Returns metrics to evaluate reconstruction quality.
    '''
    metrics = {
        'reactions': len(model.reactions),
        'metabolites': len(model.metabolites),
        'genes': len(model.genes),
        'gene_reaction_ratio': len(model.reactions) / max(1, len(model.genes))
    }

    # Count reaction types
    metrics['exchanges'] = len(model.exchanges)
    metrics['transport'] = len([r for r in model.reactions if 'transport' in r.name.lower()])

    # Test growth
    sol = model.optimize()
    metrics['can_grow'] = sol.status == 'optimal' and sol.objective_value > 0.001

    # Gene-reaction rules
    metrics['orphan_reactions'] = len([r for r in model.reactions if not r.genes])

    return metrics

Multiple Genome Reconstruction

import os
from pathlib import Path

def batch_reconstruction(fasta_dir, output_dir, media='M9'):
    '''Reconstruct models for multiple genomes

    Use for comparative genomics or community modeling.
    '''
    os.makedirs(output_dir, exist_ok=True)

    for fasta in Path(fasta_dir).glob('*.faa'):
        output = Path(output_dir) / f'{fasta.stem}.xml'
        reconstruct_model(str(fasta), str(output), media=media)
        print(f'Completed: {fasta.name}')

Community Model Construction

def merge_models(model_paths, community_name='community'):
    '''Create community model from individual organisms

    For microbiome FBA, need to create a shared compartment
    for metabolite exchange between organisms.
    '''
    import cobra

    models = [cobra.io.read_sbml_model(p) for p in model_paths]

    # Add species prefix to all components
    for i, model in enumerate(models):
        species_id = f'sp{i+1}'
        for rxn in model.reactions:
            rxn.id = f'{species_id}_{rxn.id}'
        for met in model.metabolites:
            met.id = f'{species_id}_{met.id}'
        for gene in model.genes:
            gene.id = f'{species_id}_{gene.id}'

    # Merge into community model
    community = models[0].copy()
    for model in models[1:]:
        community.merge(model)

    return community

Related Skills

  • systems-biology/model-curation - Validate and curate draft models
  • systems-biology/flux-balance-analysis - Analyze reconstructed models
  • database-access/entrez-fetch - Download genome sequences

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.