What is CultureBotAI?

CultureBotAI is a research initiative at Lawrence Berkeley National Laboratory in Berkeley, California, that develops AI-driven tools and knowledge graphs for microbial cultivation and computational biology.

Who leads CultureBotAI?

CultureBotAI is led by Dr. Marcin P. Joachimiak, a scientist specializing in microbiology, knowledge graph development, and computational biology at Lawrence Berkeley National Laboratory.

Where is CultureBotAI based?

CultureBotAI is based at Lawrence Berkeley National Laboratory in Berkeley, California, within the Environmental Genomics and Systems Biology Division.

KG-Microbe is a comprehensive modular knowledge graph developed by Dr. Marcin P. Joachimiak that integrates diverse microbial data sources to enable AI-driven insights for growth prediction and culture optimization.

How can I access KG-Microbe?

KG-Microbe is available on GitHub at https://github.com/Knowledge-Graph-Hub/kg-microbe under the BSD-3-Clause license. The publication is available in GigaScience at https://doi.org/10.1093/gigascience/giag077

What does CultureBotAI work on?

CultureBotAI focuses on three main areas: (1) cultivation of isolated and novel organisms, (2) culture optimization through data-driven approaches, and (3) growth preference prediction using machine learning and AI methods.

CultureMech: Autonomous Knowledge Factory for Culture Media

Overview

CultureMech is an autonomous knowledge factory for microbial culture media, combining ontology grounding, validation, provenance, and recipe deduplication with human oversight. Its current repository holds 15,910 normalized records and 6,320 merged canonical media. Normalized records preserve each source’s formulation; merged records provide the deduplicated view. Repository snapshot.

Explore the Published Site

Recipe browser — search source-specific recipes and filter by category, medium type, physical state, target organism, and ingredients. Its generated browser data is published by the repository’s GitHub Actions workflow.
Ingredient-derived media map — explore media using aggregated ingredient embeddings.
Direct media map — explore the medium nodes’ own KG-Microbe embeddings.
Graph layout — another view of media similarity.

Counts were checked on October 5, 2026 against the repository’s generated corpus inventory. The landing page now distinguishes 15,910 normalized records from 6,320 merged records.

Data Architecture

Layer	Location	Purpose
Raw sources	`data/raw/`	Original source payloads
Raw YAML	`data/raw_yaml/`	Source-shaped conversions
Normalized records	`data/normalized_yaml/`	Authoritative source-specific curation and browser input
Merged records	`data/merge_yaml/merged/`	Reproducible deduplicated media

Source imports include MediaDive/DSMZ, TogoMedium, KOMODO, and collection-specific recipes. The normalized inventory contains 14,329 bacterial, 746 archaeal, 254 algal, 126 fungal, and 455 specialized records. These categories sum to the normalized total, not the canonical total. Data layers.

Curation and Validation

Records carry ingredient amounts, ontology identifiers and labels, source references, and curation history. The LinkML schema separates composition type, nutritional class, and functional role while retaining the compatibility medium_type field. Validation checks schema shape, stricter record invariants, and recipe identifiers.

MediaIngredientMech supplies ingredient identity and mapping artifacts. CultureMech retains recipe-specific composition and provenance. These structured outputs support downstream knowledge-graph integration, comparative media analysis, and cultivation research. Schema · Curation guide.

Example: A Tracked Recipe

The repository’s LB medium record has the stable identifier CultureMech:008037 and preserves its TogoMedium/NBRC source. Its sodium-chloride ingredient illustrates the actual nested record format:

preferred_term: NaCl
concentration:
  value: '5'
  unit: G_PER_L
source: NBRC Medium 275
term:
  id: CHEBI:26710
  label: sodium chloride

This is an ingredient excerpt; the complete recipe also contains other ingredients, medium classification, source details, and history.

Getting Started

Development and CI use Python 3.13, uv, and just. The current checkout workflow is:

git clone https://github.com/CultureBotAI/CultureMech.git
cd CultureMech
uv sync --frozen --extra dev
just validate-schema data/normalized_yaml/bacterial/lb_medium.yaml
just test-fast

To generate local outputs:

just build-browser
just gen-pages
just gen-media-pages
just serve-browser

For one recipe, use just gen-page data/normalized_yaml/bacterial/lb_medium.yaml; the result is written under pages/single/. See the current quick start for the full workflow.

Repository & Documentation

Repository and published site
Recipe identifier lifecycle
Citation metadata
Licensing policy: CC BY 4.0 for data; BSD-3-Clause for code. Redistributed source material retains its applicable terms.

X-Mech Suite overview - All twelve Mechs, their shared vocabulary and cross-references, and the culturebotai-claw orchestrator
HabitatMech - Habitats harmonized from GOLD, BacDive, PREGO and Madin et al. into ontology-grounded records
CommunityMech - Microbial community interaction modeling
TaxonMech - Microbial taxa and strains grounded in NCBI Taxonomy, harmonized with GTDB, LPSN and BacDive
TraitMech - Autonomous knowledge factory for microbial ecophysiological traits
PathwayMech - Microbial pathway mechanisms with participants, reactions, causal edges, and reference-backed evidence
CellStructureMech - Microbial cell structures, between the trait and protein layers
ProteinTraitsMech - Protein sequence, structure, and function traits
DUFMech - Domains and protein families of unknown function, with evidence gathering and triage worklists
NaturalProductMech - Natural product structures with their producer organisms and gene clusters
AntibioticMech - Antimicrobial structures harmonizing ChEBI and CARD/ARO
MediaIngredientMech - LLM-assisted ingredient ontology mapping
MicroMediaParam - Chemical compound standardization and ChEBI mapping
kg-microbe - Central knowledge graph for microbial cultivation

Research Impact

CultureMech is part of the KG-Microbe knowledge graph ecosystem developed at Lawrence Berkeley National Laboratory. It supports:

Data-driven cultivation optimization
AI-powered growth prediction
Systematic analysis of microbial growth requirements
Evidence-based media design for novel organisms

Citation: See the KG-Microbe publication in GigaScience for details on the broader knowledge graph ecosystem.

Contact & Collaboration

For questions about CultureMech or collaboration opportunities:

Principal Investigator: Dr. Marcin P. Joachimiak
Email: mjoachimiak@lbl.gov
Organization: CultureBotAI
Laboratory: Environmental Genomics and Systems Biology Division, Lawrence Berkeley National Laboratory

Bibliography

Santangelo BE, Hegde H, Caufield JH, Reese J, Kliegr T, Hunter LE, Lozupone CA, Mungall CJ, Joachimiak MP. KG-Microbe — Building Modular and Scalable Knowledge Graphs for Microbiome and Microbial Sciences. GigaScience. 2026;giag077. doi:10.1093/gigascience/giag077
Máša P, Kliegr T, Joachimiak MP. Explainable rule-based prediction of cultivation media for microbes. Computational and Structural Biotechnology Journal. 2025;27:5194–5206. doi:10.1016/j.csbj.2025.10.014 · free full text
METPO: Microbial Ecophysiological Trait and Phenotype Ontology. BioPortal · GitHub

Full publication list →