mining-experiment-protocol
Organize the protocol around the question "What exactly can this set of experiments answer?", ensuring every group, every measurement point, and every replicate has a clear purpose.
Core Capabilities
- Select design based on research objective: Distinguish between factor screening, mechanism verification, interaction analysis, and parameter optimization, and choose the corresponding experimental arrangement.
- Utilize limited experimental resources: Arrange grouping and execution order based on specimen, equipment, and site conditions; identify replicates that can be reduced and controls that cannot be omitted.
- Detect design flaws early: Check for batch confounding, pseudo-replication, insufficient measurements, and infeasible factor combinations; correct problems before data collection.
Getting Started
From existing materials, determine the research question, primary response variable, controllable factors, experimental objects, and resource constraints. If an existing protocol is available, prioritize targeted improvements; if only research ideas exist, first propose a suitable design and justification. Only ask for information that would change the design choice; for any remaining gaps, proceed with clear assumptions.
Workflow
- Clarify the research comparison, independent experimental units, and primary response variable.
- Select the design based on factor count, interactions, batch and resource constraints; invoke a local script when a full factorial layout is needed.
- Translate experimental combinations into measurements, execution order, and analysis methods; explain what questions the design can and cannot answer.
Use Resources as Needed
- Before using local tools, read Parameters and Scope, then call scripts/generate_design.py. The tool only handles the specified format and capabilities; it does not represent completion of the full task.
- When first learning about the input format, review the synthetic example examples/design.json; demo data must not be treated as actual measurements or specification parameters.
- For method selection, boundary judgment, or complex scenarios: read references/methods.md.
- When you need to quickly organize deliverable content, copy and trim assets/protocol-outline.md; this is an optional starting point, not a fixed output format.
Deliverable
Provide the recommended design, key selection rationale, and executable experimental arrangement; when comparing designs, explain trade-offs. Attach factor combinations, measurement layouts, or analysis ideas as appropriate to the task, using the user's existing format—no fixed number of tables or section length.
Experimental parameters and field implementation conditions must be justified. Do not write results for unfinished experiments; for field operations, specific execution must conform to applicable approved plans.