SEEK v1.0¶

The full data model FAIRDOM-SEEK can represent. Use it to explore and visualize what a SEEK instance can store — it is a reference model, not a profile to build uploadable datasets on. Only its ISA core (Investigation, Study, Assay, Sample) syncs through the SEEK export, and its Project root does not, so a dataset built on it would upload only in part. To build metadata you will publish to a SEEK instance, start from the seek-ready-template profile, which is shaped so every entity uploads.

It is built on the Just Enough Results Model (JERM), the ontology SEEK types its resources with, and extends the ISA (Investigation-Study-Assay) structure with asset types for computational models, workflows, and collaborative project management. The profile is named for the platform rather than the ontology, because it is the platform you are describing data for; jerm: remains the ontology namespace the export emits.

Overview¶

JERM organizes research data hierarchically:

Project: Top-level organizational container for collaborative research
Investigation: High-level research context and goals
Study: Series of experiments addressing a biological question
Assay: Individual experimental or computational analysis

Assets can be associated at various levels:

DataFile: Experimental data in any format
Model: Computational/mathematical models (SBML, CellML, etc.)
SOP: Standard Operating Procedures
Document: Reports, specifications, protocols
Presentation: Slides, posters
Workflow: Computational pipelines (Galaxy, CWL, Nextflow)

Entity Diagram¶

flowchart TB
    subgraph project["Project Management"]
        PRJ[Project]
        INST[Institution]
    end

    subgraph isa["ISA Structure"]
        INV[Investigation]
        STU[Study]
        ASS[Assay]
    end

    subgraph assets["Assets"]
        DF[DataFile]
        MOD[Model]
        SOP[SOP]
        DOC[Document]
        PRES[Presentation]
        WF[Workflow]
        COLL[Collection]
    end

    subgraph bio["Biological"]
        SAM[Sample]
        ORG[Organism]
        STR[Strain]
    end

    subgraph people["People & Events"]
        PER[Person]
        PUB[Publication]
        EVT[Event]
    end

    PRJ --> INST
    PRJ --> INV
    PRJ --> PER

    INV --> STU
    INV --> PUB

    STU --> ASS
    STU --> PUB

    ASS --> DF
    ASS --> MOD
    ASS --> SOP
    ASS --> DOC
    ASS --> SAM

    SAM --> ORG
    SAM --> STR
    ORG --> STR

    PRES --> EVT

    MOD --> DF
    MOD --> ORG

    COLL --> DF
    COLL --> MOD
    COLL --> DOC

    classDef proj fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    classDef isa fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
    classDef asset fill:#fff3e0,stroke:#f57c00,stroke-width:2px
    classDef bio fill:#fce4ec,stroke:#c2185b,stroke-width:2px
    classDef people fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px

    class PRJ,INST proj
    class INV,STU,ASS isa
    class DF,MOD,SOP,DOC,PRES,WF,COLL asset
    class SAM,ORG,STR bio
    class PER,PUB,EVT people

Hold "Alt" / "Option" to enable pan & zoom

Entities¶

Category	Entities
Project	Project, Institution
ISA Core	Investigation, Study, Assay
Assets	DataFile, Model, SOP, Document, Presentation, Workflow, Collection, CollectionItem
Biological	Sample, SampleType, Organism, Strain
People	Person, Publication, Event
Annotations	OntologyAnnotation, Factor

Key Differences from ISA¶

Aspect	ISA	JERM
Top-level	Investigation	Project
Models	Not included	First-class Model entity
Protocols	Protocol entity	SOP entity
Workflows	Not included	Workflow entity
Collections	Not included	Collection for grouping assets
Organisms	Via OntologyAnnotation	Dedicated Organism/Strain entities
Events	Not included	Event entity for conferences

Model Formats¶

JERM supports computational models in various formats:

Format	Description
SBML	Systems Biology Markup Language
CellML	Cell modeling language
BioPAX	Biological Pathway Exchange
MATLAB	MATLAB scripts and functions
Python	Python scripts
R	R scripts

Sample Types¶

JERM uses a flexible sample type system where:

SampleType defines the template with attribute definitions
SampleAttributeDefinition specifies required/optional attributes
Sample instances conform to their type with SampleAttribute values

This allows custom sample types for different experimental domains.

SEEK Integration¶

JERM is the underlying model for FAIRDOM-SEEK. Key integration points:

SEEK IDs: Persistent URIs for all entities (seek_id field)
Versioning: Assets track version numbers
Licensing: Default and per-asset license information
Sharing policies: Project-level default policies

References¶

Resource	URL
JERM Ontology	https://jermontology.org/
FAIRDOM-SEEK	https://seek4science.org/
FAIRDOMHub	https://fairdomhub.org/
SEEK API	https://docs.seek4science.org/help/user-guide/api.html
JERM GitHub	https://github.com/FAIRdom/JERMOntology

Usage¶

from metaseed.specs.loader import SpecLoader

loader = SpecLoader()
seek = loader.load_profile(version="1.0", profile="seek")

# List all entities
print(seek.list_entities())

# Get specific entity
project = seek.get_entity("Project")
for field in project.fields:
    print(f"{field.name}: {field.type.value}")