Assay Optimization World

A laboratory World for improving a screening assay: plates, reagents, instruments, experiment runs, and measurements.

Scope

Included: assay, plate, reagent, instrument, experiment_run, measurement, protocol.

Excluded: lims_implementation, instrument_control_software.

Sources

• Ontology: openworld-examples/lab-ontology@0.1.0, a dependency.
• Knowledge graph: knowledge/plate-graph.yaml (kg/plates.ttl), an illustrative plate layout graph.
• Procedures: knowledge/sop-library.yaml, which points at docs/plate-handling-sop.md (illustrative).

Use it for

• Does the assay separate positive and negative controls well enough (eval/assay-validation.yaml: Z-factor threshold 0.5, control CV threshold 10%)?
• Should the assay be requalified when the detection antibody lot changes (scenarios/reagent-change.yaml)?
• Which wells on a campaign's plates are controls, and which reagent lot did each use (extraction/campaign-wells.yaml)?
• Who may propose experiment runs, and who approves protocols (delegations/plan-runs.yaml)?

Limitations

• Illustrative example; no real laboratory data. Uses experimental asset kinds (spec Appendix C).
• Declared profile viewable: the package has World Views but no State Compiler.

Versions

• 0.1.0: first public example.

Work

The principal investigator (a person) designs and approves protocols. A planning agent proposes experiment runs under a time-limited delegation and cannot approve protocols. A lab robot executes runs. The work follows a design → run → measure → evaluate loop (patterns/design-test-learn.yaml) and produces a protocol and a results table. eval/assay-validation.yaml validates the assay with a Z-factor threshold, and scenarios/reagent-change.yaml asks what happens if a reagent lot changes.

All task, work, artifact, actor, and scenario assets are experimental (spec Appendix C). The data is illustrative.

Assets

Path	Kind
actors/lab-robot.yaml	ActorProfile
actors/planning-agent.yaml	ActorProfile
actors/scientist.yaml	ActorProfile
agents/planner.yaml	AgentProfile
artifacts/protocol.yaml	ArtifactContract
artifacts/results-table.yaml	ArtifactContract
capabilities/assay-design.yaml	CapabilityContract
consumers/planning-agent.yaml	ConsumerRepresentationProfile
delegations/plan-runs.yaml	DelegationProfile
eval/assay-validation.yaml	EvaluationProfile
examples/baseline-observations.yaml	PackageExample
extraction/campaign-wells.yaml	KnowledgeExtractionProfile
knowledge/plate-graph.yaml	KnowledgeAsset
knowledge/sop-library.yaml	KnowledgeAsset
patterns/design-test-learn.yaml	WorkPatternProfile
roles/principal-investigator.yaml	RoleProfile
scenarios/reagent-change.yaml	ScenarioProfile
tasks/optimize-assay.yaml	TaskSetProfile
views/assay-base.yaml	WorldViewProfile
views/scientist.yaml	WorldViewProfile