fix: ONT amplicon pipeline uses dedicated ont_amplicon_params config (#79)

[berntpopp]

Summary

The ONT amplicon pipeline was reading model_file from pacbio_params, which defaults to ERRHMM-SEQUEL.model (a PacBio Sequel error model). ONT reads were being simulated with PacBio error profiles. The v0.43.5 "fix" only added a runtime warning but still used the wrong model.

Root cause

No ont_amplicon_params config section existed. The ONT pipeline shared pacbio_params with PacBio, and the CLI wrote --model-file overrides to pacbio_params for both platforms.

Fix

• Add ont_amplicon_params config section with ERRHMM-ONT.model as default
• Add OntAmpliconConfig TypedDict in type_defs.py
• ONT pipeline reads from ont_amplicon_params instead of pacbio_params
• CLI routes --model-file to ont_amplicon_params for --platform ont, pacbio_params for PacBio
• Config schema validation and path resolution for ont_amplicon_params

Config change

"ont_amplicon_params": {
    "model_type": "errhmm",
    "model_file": "reference/pbsim3/ERRHMM-ONT.model",
    "threads": 8,
    "accuracy_mean": 0.95
}

Test plan

• 1,431 tests pass (89% coverage)
• ruff clean, mypy clean
• New test: ONT platform populates ont_amplicon_params (not pacbio_params)
• Existing ONT pipeline tests updated to use ont_amplicon_params
• Documentation updated (amplicon simulation guide, changelog)
• CI passes

Closes #79

🤖 Generated with Claude Code

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 69.69697% with 10 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
muc_one_up/cli/commands/reads.py	47.36%	4 Missing and 6 partials ⚠️

📢 Thoughts on this report? Let us know!

[Copilot]

Pull request overview

This PR fixes ONT amplicon simulation incorrectly using PacBio error-model configuration by introducing a dedicated ont_amplicon_params config section and routing ONT-specific CLI/pipeline behavior through it.

Changes:

• Added ont_amplicon_params to the config schema/default config, plus a new OntAmpliconConfig TypedDict.
• Updated the ONT amplicon pipeline to read model settings from ont_amplicon_params (with ONT defaults) instead of pacbio_params.
• Updated CLI and tests/docs so --platform ont populates/uses ont_amplicon_params.

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
tests/read_simulator/test_ont_amplicon_pipeline.py	Updates fixtures/tests to use ont_amplicon_params instead of pacbio_params.
tests/cli/test_amplicon_platform.py	Adds CLI test asserting ONT platform populates ont_amplicon_params.
muc_one_up/type_defs.py	Adds OntAmpliconConfig TypedDict for ONT amplicon params.
muc_one_up/read_simulator/ont_amplicon_pipeline.py	Switches ONT pipeline parameter source to ont_amplicon_params and adjusts coverage/model handling.
muc_one_up/config.py	Adds JSON schema + path resolution support for ont_amplicon_params.model_file.
muc_one_up/cli/commands/reads.py	Routes --model-file/--model-type/--seed overrides to ont_amplicon_params when --platform ont.
docs/guides/amplicon-simulation.md	Documents the new ONT config section and defaults.
docs/about/changelog.md	Adds changelog entries for the fix/new config section.
config.json	Adds default ont_amplicon_params section with ONT model defaults.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

ont_params.get(...) will return None if the key exists but is explicitly set to null (possible when using load_config_raw, which skips schema validation). That would later break Path(model_file) and also pass None into run_pbsim3_template_simulation where accuracy_mean is formatted as a float. Consider normalizing here (e.g., treat None as missing and fall back to defaults) or raise a clear error if model_type/model_file/accuracy_mean are present but null/invalid.

[Copilot]

The defaults here only apply when the key is absent, not when it is present-but-null (e.g., {"ont_amplicon_params": {"model_file": null}} from a partial config loaded via load_config_raw). In that case the pipeline will later crash when it tries to use None as a path/string. Consider treating None the same as missing (or validating and raising a click.ClickException), similar to how _apply_pacbio_params validates required params.

Suggested change

	# Default to ERRHMM-ONT.model if no model configured
	if "model_file" not in params:
	params["model_file"] = "reference/pbsim3/ERRHMM-ONT.model"
	logging.info("Using default ONT model: %s", params["model_file"])
	if "model_type" not in params:
	# Default to ERRHMM-ONT.model if no model configured. Treat explicit
	# nulls from partially loaded config the same as missing values.
	if params.get("model_file") is None:
	params["model_file"] = "reference/pbsim3/ERRHMM-ONT.model"
	logging.info("Using default ONT model: %s", params["model_file"])
	if params.get("model_type") is None:

[Copilot]

This docstring says _apply_ont_amplicon_params raises click.ClickException when required params are missing, but the function currently never raises (it silently fills defaults). Either remove/update the "Raises" section, or add explicit validation for invalid/None values so callers get a consistent, actionable CLI error (matching _apply_pacbio_params behavior).

Suggested change

	Raises:
	click.ClickException: If required params are missing after merge.
	Missing model settings are populated with defaults if not provided.

[Copilot]

accuracy_mean is allowed to be null by the schema, but the ONT amplicon pipeline passes this value directly into run_pbsim3_template_simulation, which expects a float and formats it with %.2f. Either disallow null here (use just "type": "number") or ensure the pipeline coerces null to the default before invoking pbsim3.

Suggested change

	"type": ["number", "null"],
	"type": "number",