CommunityMech: Microbial Community Mechanisms Knowledge Base

Overview

CommunityMech is a LinkML-based knowledge base for modeling microbial community interactions and multi-organism cultivation systems. It provides structured representation of community composition, ecological interactions (syntrophy, competition, facilitation), and evidence-based relationship tracking to support consortium design and polyculture optimization.

The Challenge: Microbial communities exhibit complex ecological interactions that are critical for successful cultivation, yet these relationships are scattered across literature in unstructured formats. Single-organism cultivation data cannot capture the syntrophic dependencies, competitive exclusions, and facilitative interactions that determine community assembly and stability.

The Solution: CommunityMech provides a structured schema for representing community-level cultivation data, interaction networks, and temporal dynamics. It extends the single-organism focus of CultureMech to multi-species systems, enabling systematic analysis of community cultivation requirements.

Current Contents

The knowledge base holds 304 curated communities across 16 categories, each with evidence-backed interaction networks and 512-dimensional (v3) embeddings for similarity search.

Browse it through the record browser or explore the PaCMAP embedding map of the community space.

Key Features

๐Ÿ”ฌ LinkML Schema Design

๐ŸŒ Ecological Interactions Framework

๐Ÿ“š Evidence-Based Tracking

๐Ÿงฌ Community Assembly Modeling

๐Ÿ“Š Network Visualization

๐Ÿ’พ Multi-Format Export

Technical Architecture

Community Modeling Pipeline

Literature & Experimental Data
    โ†“
Community Composition Extraction
    โ†“
Interaction Identification
    โ†“
Evidence Classification
    โ†“
LinkML Schema Validation
    โ†“
CommunityMech Knowledge Base
    โ†“
โ”œโ”€ Consortium Design (PFASCommunityAgents)
โ”œโ”€ Growth Predictions (Multi-organism models)
โ””โ”€ Community Analysis (Network metrics)

Integration with CultureBotAI Ecosystem

CommunityMech extends single-organism cultivation data to multi-species systems:

Ecological Interactions Framework

Positive Interactions

Syntrophy

Obligate metabolic cooperation where one organism depends on metabolites produced by another.

Example: Syntrophus species and methanogenic archaea

{
  "interaction_type": "syntrophy",
  "partner_1": {"taxon": "Syntrophus aciditrophicus", "role": "fatty acid oxidizer"},
  "partner_2": {"taxon": "Methanospirillum hungatei", "role": "H2 consumer"},
  "mechanism": "Interspecies hydrogen transfer",
  "evidence": "PMID:12345678",
  "context": "Anaerobic conditions, <10 ยตM H2"
}

Mutualism

Both organisms benefit from the interaction.

Commensalism

One organism benefits, the other is unaffected.

Facilitation

One organism modifies the environment to benefit another (e.g., O2 removal for anaerobes).

Negative Interactions

Competition

Organisms compete for the same resources.

Example: Carbon source competition

{
  "interaction_type": "competition",
  "partner_1": {"taxon": "Escherichia coli"},
  "partner_2": {"taxon": "Salmonella enterica"},
  "mechanism": "Glucose uptake competition",
  "outcome": "Competitive exclusion at low glucose",
  "evidence": "PMID:87654321"
}

Inhibition

One organism produces compounds that inhibit another (antibiotics, bacteriocins, pH changes).

Context-Dependent Interactions

Many interactions depend on environmental conditions:

{
  "interaction_type": "context_dependent",
  "partner_1": {"taxon": "Pseudomonas putida"},
  "partner_2": {"taxon": "Bacillus subtilis"},
  "conditions": [
    {"context": "High nutrient", "interaction": "competition"},
    {"context": "Low nutrient", "interaction": "facilitation",
     "mechanism": "P. putida exopolysaccharide aids B. subtilis biofilm"}
  ]
}

LinkML Schema

CommunityMech uses LinkML (Linked Data Modeling Language) for structured, validated data representation:

Core Classes

classes:
  MicrobialCommunity:
    description: A collection of microbial species cultivated together
    attributes:
      community_id:
        range: string
        required: true
      members:
        range: CommunityMember
        multivalued: true
      interactions:
        range: Interaction
        multivalued: true
      cultivation_conditions:
        range: CultivationConditions

  CommunityMember:
    description: A microbial species within a community
    attributes:
      taxon:
        range: string
        description: NCBI taxonomy ID or species name
      abundance:
        range: float
        description: Relative abundance (0-1)
      role:
        range: FunctionalRole
        description: Primary/secondary producer, degrader, etc.

  Interaction:
    description: Ecological interaction between community members
    attributes:
      interaction_type:
        range: InteractionType
        required: true
      partner_1:
        range: CommunityMember
      partner_2:
        range: CommunityMember
      mechanism:
        range: string
        description: Molecular mechanism (if known)
      evidence:
        range: Evidence
        multivalued: true
      confidence:
        range: float
        description: Confidence score (0-1)

Use Cases

1. PFAS Biodegradation Consortia Design

Design optimized microbial consortia for PFAS (per- and polyfluoroalkyl substances) biodegradation:

{
  "community_name": "PFAS-degrading consortium v2.1",
  "target_compound": "PFOA (perfluorooctanoic acid)",
  "members": [
    {
      "taxon": "Acidimicrobium sp. A6",
      "role": "PFOA defluorination",
      "evidence": "Genomic capacity (pfa gene cluster)"
    },
    {
      "taxon": "Pseudomonas sp. B12",
      "role": "Short-chain intermediate degradation",
      "evidence": "Growth on fluoroacetate"
    },
    {
      "taxon": "Rhizobium sp. C3",
      "role": "Nitrogen fixation support",
      "evidence": "nif genes, enhances P. sp. B12 growth"
    }
  ],
  "interactions": [
    {
      "type": "syntrophy",
      "partners": ["A6", "B12"],
      "mechanism": "A6 defluorinates PFOA โ†’ B12 degrades short-chain products"
    },
    {
      "type": "facilitation",
      "partners": ["C3", "B12"],
      "mechanism": "N2 fixation supports B12 under N-limited conditions"
    }
  ]
}

2. Syntrophic Community Engineering

Model obligate syntrophic partnerships for anaerobic digestion or specialized metabolism:

3. Microbiome Cultivation

Optimize cultivation of complex microbiomes from environmental samples:

4. Community Stability Prediction

Predict which community compositions will remain stable under specific conditions:

5. Bioaugmentation Strategy

Design bioaugmentation strategies by modeling how introduced species will interact with existing communities.

Example: Methanogenic Consortium

Community Definition

{
  "community_id": "METHAN-01",
  "name": "Anaerobic digester methanogenic consortium",
  "members": [
    {
      "taxon_id": "NCBI:572546",
      "name": "Syntrophus aciditrophicus",
      "role": "Fatty acid oxidizer",
      "abundance": 0.15
    },
    {
      "taxon_id": "NCBI:879972",
      "name": "Methanospirillum hungatei",
      "role": "Hydrogenotrophic methanogen",
      "abundance": 0.25
    },
    {
      "taxon_id": "NCBI:2159",
      "name": "Methanosarcina barkeri",
      "role": "Acetoclastic methanogen",
      "abundance": 0.20
    },
    {
      "taxon_id": "NCBI:1736",
      "name": "Clostridium cellulolyticum",
      "role": "Cellulose degrader",
      "abundance": 0.40
    }
  ],
  "interactions": [
    {
      "type": "syntrophy",
      "partner_1": "Syntrophus aciditrophicus",
      "partner_2": "Methanospirillum hungatei",
      "mechanism": "Interspecies hydrogen transfer",
      "metabolites": ["H2", "formate"],
      "evidence": {
        "type": "co-culture",
        "reference": "PMID:12345678",
        "confidence": 0.95
      }
    },
    {
      "type": "facilitation",
      "partner_1": "Clostridium cellulolyticum",
      "partner_2": "Syntrophus aciditrophicus",
      "mechanism": "Cellulose โ†’ short-chain fatty acids",
      "metabolites": ["acetate", "butyrate", "propionate"],
      "evidence": {
        "type": "metabolomics",
        "reference": "PMID:23456789",
        "confidence": 0.90
      }
    },
    {
      "type": "competition",
      "partner_1": "Methanospirillum hungatei",
      "partner_2": "Methanosarcina barkeri",
      "mechanism": "Acetate utilization",
      "context": "High acetate concentrations",
      "outcome": "M. barkeri outcompetes at >5 mM acetate",
      "evidence": {
        "type": "competition assay",
        "reference": "PMID:34567890",
        "confidence": 0.85
      }
    }
  ],
  "cultivation_conditions": {
    "temperature": 37,
    "ph": 7.2,
    "atmosphere": "Anaerobic (N2/CO2, 80:20)",
    "media": "DSMZ Medium 641 (Methanogenic Medium)"
  }
}

Network Visualization

    [C. cellulolyticum] --(SCFAs)--> [S. aciditrophicus]
                                            |
                                        (H2, formate)
                                            โ†“
                                   [M. hungatei] -----> CH4
                                            |
                                      (competition)
                                            โ†“
                                   [M. barkeri] ------> CH4

Repository & Documentation

Topics

microbes ยท microbial-communities ยท microbial-community-assembly ยท microbial-ecology ยท microbial-interactions ยท microbiology ยท microbiome ยท microbiome-analysis ยท microbiome-data ยท microbial-community-dynamics

Getting Started

Installation

# Clone the repository
git clone https://github.com/CultureBotAI/CommunityMech.git
cd CommunityMech

# Install dependencies
pip install -r requirements.txt

# Validate LinkML schema
linkml-validate --schema schema/community_mech.yaml data/example_community.json

Basic Usage

from communitymech import CommunityModel

# Load a community from data
community = CommunityModel.load("data/methanogenic_consortium.json")

# Access community members
for member in community.members:
    print(f"{member.name}: {member.role} ({member.abundance:.2%})")

# Query interactions
syntrophic = community.get_interactions(type="syntrophy")
for interaction in syntrophic:
    print(f"{interaction.partner_1} <-> {interaction.partner_2}")
    print(f"  Mechanism: {interaction.mechanism}")

# Export network for visualization
community.export_network("community_network.graphml")

# Validate community composition
validation = community.validate()
print(f"Validation status: {validation.status}")

Integration with PFASCommunityAgents

CommunityMech provides the knowledge base for PFASCommunityAgents, an AI-driven consortium design system:

from pfascommunityagents import ConsortiumDesigner
from communitymech import CommunityModel

# Load existing community knowledge
knowledge_base = CommunityModel.load_knowledge_base()

# Design consortium for PFOA degradation
designer = ConsortiumDesigner(knowledge_base)
consortium = designer.design_for_target(
    compound="PFOA",
    objectives=["complete_mineralization", "stability", "rapid_growth"],
    constraints=["aerobic_conditions", "temperature_range_20_30C"]
)

# Export designed consortium
consortium.save("pfoa_consortium_v1.json")

Research Impact

CommunityMech enables systematic study of microbial community cultivation by:

It is part of the KG-Microbe knowledge graph ecosystem at Lawrence Berkeley National Laboratory, supporting research in environmental biotechnology, bioremediation, and microbial ecology.

Publications & Citations

Primary Citation

If you use CommunityMech in your research, please cite:

Santangelo BE, et al. (2026). KG-Microbe - Building Modular and Scalable Knowledge Graphs for Microbiome and Microbial Sciences. GigaScience, giag077. doi: 10.1093/gigascience/giag077

Future Directions

Planned Features

Community Contributions

We welcome contributions for:

Contact & Collaboration

For questions about CommunityMech or consortium design projects:

Acknowledgments

CommunityMech development is supported by:

Bibliography

  1. Santangelo BE, Hegde H, Caufield JH, Reese J, Kliegr T, Hunter LE, Lozupone CA, Mungall CJ, Joachimiak MP. KG-Microbe โ€” Building Modular and Scalable Knowledge Graphs for Microbiome and Microbial Sciences. GigaScience. 2026;giag077. doi:10.1093/gigascience/giag077
  2. Naseem S, Miller MA, Martinez-Gomez NC, Sun N, Joachimiak MP. MicroGrowAgents: An Agentic AI System for Microbial Cultivation Engineering. bioRxiv. 2026. doi:10.64898/2026.06.04.729985
  3. Mรกลกa P, Kliegr T, Joachimiak MP. Explainable rule-based prediction of cultivation media for microbes. Computational and Structural Biotechnology Journal. 2025;27:5194โ€“5206. doi:10.1016/j.csbj.2025.10.014 ยท free full text

Full publication list โ†’