feat: add experimental insect hosts

[kkyungseo]

파일	역할
src/factorforge/data/spodoptera_frugiperda_codons.json	Sf9 플레이스홀더 코돈 테이블 (GC≈41%, AGA→Arg)
src/factorforge/data/trichoplusia_ni_codons.json	Tni 플레이스홀더 코돈 테이블 (GC≈39%, 더 강한 AT-bias)
src/factorforge/cli/main.py (수정)	HOST_MAP, click.Choice, help text에 sf9 / tni 추가
src/factorforge/registry/current_parameter_registry.yaml (수정)	host_profiles에 sf9, tni 추가 (status=experimental, owner_job=023)
tests/engines/profile/test_host_insects.py	18개 테스트 추가 및 통과, 기존 494개 테스트 무회귀 확인

플레이스홀더 설계 원칙

• total_cds=0 값을 사용하여 플레이스홀더 상태를 명시

• source 필드에 "PLACEHOLDER" 포함 → 테스트에서 강제 검증

• 레피도프테라(Lepidoptera) 특성 보존

  • AGA → Arg 우선 사용

  • 강한 AT-bias 반영

  • GC(Tni) < GC(Sf9) 관계 유지

• 실제 데이터 확보 시 JSON 파일 교체만으로 파이프라인 완성 가능

[vercel]

@KyungSeo116 is attempting to deploy a commit to the munkyukim86's projects Team on Vercel.

A member of the Team first needs to authorize it.

[munkyukim86]

Code review

Found 3 issues:

1. New hosts added to CLI but not to the REST API — api/optimize.py still has VALID_HOSTS = ["nbenthamiana", "by2"] and its own HOST_MAP that were not updated. Any POST /optimize request using sf9 or tni will be rejected (line 274 raises ValueError: Unsupported host). Note: PR feat: add experimental plant hosts #75 has the same omission — both PRs need api/optimize.py updated before the new hosts are reachable via the web API.

factorforge-cds/src/factorforge/cli/main.py

Lines 19 to 27 in db363f5

	HOST_MAP = {
	"nbenthamiana": "nbenthamiana",
	"by2": "ntabacum",
	"sf9": "spodoptera_frugiperda",
	"tni": "trichoplusia_ni",
	}
	def _configure_stdio() -> None:

2. No runtime warning when using placeholder codon tables — The help text mentions "placeholder codon tables" but the engine loads and uses them silently at runtime. load_codon_table() in utils.py does not inspect total_cds or the source field, so a user running --host sf9 receives optimization output with no indication that the underlying codon data is unvalidated. A click.echo() warning before optimization runs (or a WARNING: in the FASTA header) would be the minimum mitigation.

factorforge-cds/src/factorforge/data/spodoptera_frugiperda_codons.json

Lines 1 to 10 in db363f5

	{
	"organism": "Spodoptera frugiperda",
	"source": "PLACEHOLDER — awaiting verified CDS dataset. Frequencies are estimates derived from Lepidoptera genome surveys (Kazusa CodonUsage Database, Spodoptera frugiperda ISE-6 assembly). Do not use for production optimization until data is validated and source is updated.",
	"description": "PLACEHOLDER codon usage table for Spodoptera frugiperda (fall armyworm). Sf9 cells are the primary insect cell line for baculovirus expression vector systems (BEVS). Shows characteristic Lepidoptera AT-bias at synonymous positions and strong AGA preference for Arg.",
	"total_cds": 0,
	"total_codons": 0,
	"codons": {
	"TTT": {
	"aa": "F",
	"frequency": 0.46,

3. GC scoring band (55–65%) silently applied to insect hosts — The default GC band is calibrated for N. benthamiana (benchmark analysis 004). Sf9 CDS GC is ~41% and T. ni is ~39%. With no host-specific ScoringConfig override, every sequence optimized for sf9 or tni that lands in its natural GC range will receive a penalized GC score and potentially a composite-score warning. The band mismatch is ~15 percentage points. A host-specific gc_min/gc_max entry (or at minimum a documented known limitation) is needed before users interpret the scores.

factorforge-cds/src/factorforge/registry/current_parameter_registry.yaml

Lines 186 to 215 in db363f5

	owner_job: "023"
	source: "NCBI Taxonomy Browser NCBITaxon:7108; codon table is PLACEHOLDER — awaiting verified CDS dataset"
	rationale: "Sf9 insect cell line (Lepidoptera); primary BEVS host. Codon table uses placeholder frequencies derived from Lepidoptera genome surveys."
	provenance: "Issue #23 insect host addition; placeholder codon table until authoritative S. frugiperda CDS build is available."
	visibility: public
	permission: publish_allowed
	tni:
	display_name: "Tni / T. ni (High Five)"
	scientific_name: "Trichoplusia ni"
	ncbi_taxonomy_id: 7111
	ncbi_taxonomy_curie: "NCBITaxon:7111"
	status: experimental
	claim_level: experimental_setting
	evidence_status: experimental
	release_status: experimental
	owner_job: "023"
	source: "NCBI Taxonomy Browser NCBITaxon:7111; codon table is PLACEHOLDER — awaiting verified CDS dataset (ref: Tnms42 genome GCF_003590095.1)"
	rationale: "Tni High Five insect cell line (BTI-Tn5B1-4, Lepidoptera); widely used for secreted protein in BEVS. Codon table uses placeholder frequencies."
	provenance: "Issue #23 insect host addition; placeholder codon table until authoritative T. ni CDS build is available."
	visibility: public
	permission: publish_allowed
	codon_reference:
	id: nbenthamiana_legacy_kazusa_sgn_v101
	species: Nicotiana benthamiana
	ncbi_taxid: 4100
	source_label: "Kazusa CodonUsage Database + SGN genome v1.0.1"
	source_status: legacy_metadata_only
	build_path_status: incomplete
	authoritative_build_script_available: false

🤖 Generated with Claude Code

_{- If this code review was useful, please react with 👍. Otherwise, react with 👎.}

[munkyukim86]

Thanks for contributing this! Really appreciate you taking the time to add the insect hosts. Left a few notes above, once those are addressed, happy to get this merged.

[kkyungseo]

Thanks for the review! I'll work through the notes above and aim to have everything addressed by June 18 (KST).
I'll ping you here once it's ready for another look.

[munkyukim86]

Can't wait, thanks!

[munkyukim86]

Thanks for the contribution! The pandas import failure here is from a stale CI snapshot (2026-06-15) — main has since fixed the benchmark module's dependency setup. Could you rebase this branch onto current main to pick that up? Should resolve the CI failures.