Skip to main content
Glama
toxcast_overview.md6.58 kB
# BioBricks ToxCast Knowledge Graph - Chat Transcript **Date:** December 11, 2025 **Topic:** Exploring the BioBricks ToxCast Knowledge Graph --- ## Conversation ### User > @biobricks-toxcast ### Assistant I can help you explore the BioBricks ToxCast knowledge graph! This resource contains data from the EPA's ToxCast program, which tests chemicals for potential toxicity. **Retrieved Endpoint Description:** BioBricks ToxCast is an open knowledge graph for EPA ToxCast high-throughput screening data. **Key Statistics:** - **3.34 million** bioassay screening results - **9,542** chemical entities - **2,205** distinct assay endpoints - Chemicals identified by EPA DSSTox Substance Identifiers (DTXSID) and InChIKeys **Data Source:** - Derived from EPA's public domain ToxCast database - Released under CC0-1.0 license - SPARQL endpoint: https://frink.apps.renci.org/biobricks-toxcast/sparql **Contact:** - Tom Luechtefeld (tom@insilica.co) - Homepage: https://github.com/biobricks-ai/biobricks-okg --- ### Knowledge Graph Schema **Main Classes:** 1. **Chemical Entity** - A molecule or compound with defined chemical structure and properties 2. **Chemical Identifier** - A unique code or number that identifies a specific chemical substance 3. **Bioassay** - A laboratory test that measures the biological or biochemical effects of a substance 4. **Chemical Structure Descriptor** - Information that describes the arrangement of atoms in a molecule 5. **Chemical Information** - Data or knowledge about chemical substances and their properties 6. **Biological Target** - A molecule in a living system that a drug or chemical substance interacts with 7. **Assay Design** - The specific method and conditions used to conduct a biological experiment **Key Relationships (Predicates):** - **Label** - A human-readable name or title for an entity - **Source** - The origin or reference for a piece of information - **Has Identifier** - Links an entity to its unique identification code - **Participates In** - An entity takes part in or is involved in a biological process - **Assay ID** - A unique identifier for a toxicology assay in the EPA CompTox database - **Has Participant** - A process involves or includes a particular entity - **Has Specified Output** - The result or product produced by a process or assay - **Has Assay Component** - The specific part or measurement within a larger bioassay - **Is About** - Indicates that information or data pertains to a particular subject - **Has Value** - Assigns a specific measurement or quantity to an entity --- ### What can I help you explore? I can help you: 1. **Find chemicals** by name, identifier, or structure 2. **Search bioassays** by target, endpoint, or design 3. **Explore chemical-activity relationships** across toxicological targets 4. **Query screening results** for specific chemicals or assays 5. **Visualize the knowledge graph structure** with diagrams What would you like to investigate in the ToxCast data? --- ### User > Create a class diagram ### Assistant I've created a comprehensive class diagram for the BioBricks ToxCast knowledge graph! **Core Classes:** - **ChemicalEntity** (9,542 entities) - The actual molecules/compounds tested - **Bioassay** (3.34M results, 2,205 endpoints) - The laboratory tests - **BiologicalTarget** - The molecules in living systems that chemicals interact with - **AssayDesign** - The experimental methods and conditions **Supporting Classes:** - **ChemicalIdentifier** - IDs like DTXSID and InChIKeys - **ChemicalStructureDescriptor** - Information about molecular structure - **ChemicalInformation** - General data about substances **Key Relationships:** - Chemicals participate in bioassays - Bioassays have chemical participants and target biological molecules - Assays produce chemical information as outputs - Chemical entities are linked to their identifiers and structure descriptors The diagram visualizes how EPA ToxCast connects chemicals to their toxicological screening data through standardized assays and biological targets. **File Created:** toxcast_class_diagram.mermaid #### Class Diagram ```mermaid classDiagram class ChemicalEntity { +label: string +source: string } class ChemicalIdentifier { +label: string +value: string } class ChemicalStructureDescriptor { +label: string +value: string } class ChemicalInformation { +label: string +source: string } class Bioassay { +label: string +assayID: string +source: string } class BiologicalTarget { +label: string } class AssayDesign { +label: string } %% Chemical Entity Relationships ChemicalEntity --> ChemicalIdentifier : has_identifier ChemicalEntity --> ChemicalStructureDescriptor : has descriptor ChemicalEntity --> Bioassay : participates_in %% Chemical Information Relationships ChemicalInformation --> ChemicalIdentifier : has_identifier ChemicalInformation --> ChemicalStructureDescriptor : has descriptor ChemicalInformation --> Bioassay : is_about %% Bioassay Relationships Bioassay --> ChemicalEntity : has_participant Bioassay --> BiologicalTarget : targets Bioassay --> AssayDesign : has_design Bioassay --> ChemicalInformation : has_specified_output %% Assay Design Relationships AssayDesign --> Bioassay : defines %% Biological Target Relationships BiologicalTarget --> ChemicalEntity : interacts_with BiologicalTarget --> ChemicalInformation : is_about ``` --- ## Use Cases BioBricks ToxCast is useful for: - **Computational toxicology research** - Analyzing chemical-bioactivity relationships - **Regulatory science** - Supporting risk assessment decisions - **Environmental health** - Understanding chemical effects on biological systems - **Drug discovery** - Identifying potential toxicity early in development - **Adverse outcome pathway analysis** - Mapping mechanisms of toxicity - **Compound prioritization** - Screening chemicals for safety concerns --- ## Technical Details **Ontologies Used:** - BioAssay Ontology (BAO) - Chemical Information Ontology (CHEMINF) - Biolink Model - Relation Ontology (RO) **Access Methods:** - SPARQL queries at https://frink.apps.renci.org/biobricks-toxcast/sparql - Triple Pattern Fragments at https://frink.apps.renci.org/ldf/biobricks-toxcast **Funding:** - NSF Award: 2333728 --- ## End of Transcript

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sbl-sdsc/mcp-proto-okn'

If you have feedback or need assistance with the MCP directory API, please join our Discord server