factorforge-cds 3.2.0__tar.gz → 3.2.1__tar.gz

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: factorforge-cds
-Version: 3.2.0
+Version: 3.2.1
 Summary: FactorForge - open-source constraint-based CDS design engine by Eijex.
 Author-email: Eijex <eijex.lab@gmail.com>
 License-Expression: AGPL-3.0-only
@@ -21,6 +21,7 @@ Requires-Dist: click>=8.0
 Requires-Dist: pydantic>=2.0
 Provides-Extra: dev
 Requires-Dist: jsonschema>=4.0; extra == "dev"
+Requires-Dist: pandas>=2.0; extra == "dev"
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
 Requires-Dist: ruff>=0.1; extra == "dev"
@@ -39,7 +40,7 @@ Dynamic: license-file
 [![PyPI](https://img.shields.io/pypi/v/factorforge-cds.svg)](https://pypi.org/project/factorforge-cds/)
 [![CI](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml/badge.svg)](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml)
 [![codecov](https://codecov.io/gh/eijex/factorforge-cds/branch/main/graph/badge.svg)](https://codecov.io/gh/eijex/factorforge-cds)
-[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20407331.svg)](https://doi.org/10.5281/zenodo.20407331)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20640931.svg)](https://doi.org/10.5281/zenodo.20640931)
 [![Web App](https://img.shields.io/badge/web-factorforge.eijex.com-brightgreen.svg)](https://factorforge.eijex.com)
 
 FactorForge performs profile-guided CDS design with CAI/GC metrics, PolyA-signal screening, and Golden Gate/MoClo-aware checks. Primary support: *N. benthamiana* (agroinfiltration). Experimental host context: Tobacco BY-2 (`--host by2`).
@@ -113,7 +114,7 @@ GNU Affero General Public License v3.0 — see [LICENSE](LICENSE).
 ## Get in Touch
 
 - **Docs** — [eijex.github.io/factorforge-cds](https://eijex.github.io/factorforge-cds/)
-- **Wet-lab Results** — Public-safe validation summaries are welcome. Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
+- **Wet-lab Results** — Public-safe validation summaries are welcome. [Share Wet-lab Results (Form)](https://docs.google.com/forms/d/e/1FAIpQLSeSx-wYvF6YwHhSPdLMl-L44frCugdm25X_eDz50OaqTD66qA/viewform) or [Share Wet-lab Results (GitHub)](https://github.com/eijex/factorforge-cds/issues/new?template=wet_lab_result.yml) (public-safe summaries only). Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
 - **GitHub Issues** — bugs, features: [github.com/eijex/factorforge-cds/issues](https://github.com/eijex/factorforge-cds/issues)
 - **Email** — eijex.lab@gmail.com
 - **FactorForge** — [factorforge.eijex.com](https://factorforge.eijex.com)

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/README.md

@@ -7,7 +7,7 @@
 [![PyPI](https://img.shields.io/pypi/v/factorforge-cds.svg)](https://pypi.org/project/factorforge-cds/)
 [![CI](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml/badge.svg)](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml)
 [![codecov](https://codecov.io/gh/eijex/factorforge-cds/branch/main/graph/badge.svg)](https://codecov.io/gh/eijex/factorforge-cds)
-[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20407331.svg)](https://doi.org/10.5281/zenodo.20407331)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20640931.svg)](https://doi.org/10.5281/zenodo.20640931)
 [![Web App](https://img.shields.io/badge/web-factorforge.eijex.com-brightgreen.svg)](https://factorforge.eijex.com)
 
 FactorForge performs profile-guided CDS design with CAI/GC metrics, PolyA-signal screening, and Golden Gate/MoClo-aware checks. Primary support: *N. benthamiana* (agroinfiltration). Experimental host context: Tobacco BY-2 (`--host by2`).
@@ -81,7 +81,7 @@ GNU Affero General Public License v3.0 — see [LICENSE](LICENSE).
 ## Get in Touch
 
 - **Docs** — [eijex.github.io/factorforge-cds](https://eijex.github.io/factorforge-cds/)
-- **Wet-lab Results** — Public-safe validation summaries are welcome. Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
+- **Wet-lab Results** — Public-safe validation summaries are welcome. [Share Wet-lab Results (Form)](https://docs.google.com/forms/d/e/1FAIpQLSeSx-wYvF6YwHhSPdLMl-L44frCugdm25X_eDz50OaqTD66qA/viewform) or [Share Wet-lab Results (GitHub)](https://github.com/eijex/factorforge-cds/issues/new?template=wet_lab_result.yml) (public-safe summaries only). Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
 - **GitHub Issues** — bugs, features: [github.com/eijex/factorforge-cds/issues](https://github.com/eijex/factorforge-cds/issues)
 - **Email** — eijex.lab@gmail.com
 - **FactorForge** — [factorforge.eijex.com](https://factorforge.eijex.com)

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "factorforge-cds"
-version = "3.2.0"
+version = "3.2.1"
 description = "FactorForge - open-source constraint-based CDS design engine by Eijex."
 readme = "README.md"
 license = "AGPL-3.0-only"
@@ -29,6 +29,7 @@ dependencies = [
 [project.optional-dependencies]
 dev = [
     "jsonschema>=4.0",
+    "pandas>=2.0",
     "pytest>=7.0",
     "pytest-cov>=4.0",
     "ruff>=0.1",

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge/__init__.py

@@ -4,7 +4,7 @@ FactorForge - Codon Optimization Platform
 profile: constraint-aware rule/profile engine
 """
 
-__version__ = "3.2.0"
+__version__ = "3.2.1"
 __author__ = "Eijex"
 
 # Auto-register engines (safe when running from source tree)

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge/engines/profile/rules/reverse_translator.py

@@ -423,7 +423,7 @@ class ReverseTranslator:
         """
         High-CAI profile: use only preferred codons
 
-        - CAI > 0.85 guaranteed
+        - Selects codons that maximize the profile golden-set CAI reference
         - No GC constraints
         """
         dna_seq: list[str] = []

factorforge_cds-3.2.1/src/factorforge/protein_risk/__init__.py

@@ -0,0 +1,5 @@
+"""Deterministic protein structural-risk annotations."""
+
+from factorforge.protein_risk.annotate import annotate_protein_risk
+
+__all__ = ["annotate_protein_risk"]

factorforge_cds-3.2.1/src/factorforge/protein_risk/annotate.py

@@ -0,0 +1,32 @@
+"""Compose protein structural-risk annotations."""
+
+from factorforge.protein_risk.risk_classifier import classify_risk
+from factorforge.protein_risk.sp_predict import predict_signal_peptide
+from factorforge.protein_risk.tm_predict import predict_tm_segments
+
+
+def annotate_protein_risk(protein_seq: str) -> dict:
+    """Return deterministic structural-risk indicators without sequence material."""
+    tm_prediction = predict_tm_segments(protein_seq)
+    sp_predicted = predict_signal_peptide(protein_seq)
+    risk_level = classify_risk(
+        tm_prediction["tm_count"],
+        tm_prediction["mean_kd_score"],
+        sp_predicted,
+    )
+
+    warnings = []
+    if risk_level != "LOW":
+        warnings.append(
+            "Protein structural risk indicator: "
+            f"{risk_level} (tm_count={tm_prediction['tm_count']}). "
+            "CDS-level checks are independent of protein structural risk."
+        )
+
+    return {
+        "tm_count": tm_prediction["tm_count"],
+        "mean_kd_score": tm_prediction["mean_kd_score"],
+        "signal_peptide_predicted": sp_predicted,
+        "risk_level": risk_level,
+        "warnings": warnings,
+    }

factorforge_cds-3.2.1/src/factorforge/protein_risk/kd_scale.py

@@ -0,0 +1,30 @@
+"""Kyte-Doolittle scale and thresholds used by protein-risk heuristics."""
+
+KD_SCALE = {
+    "A": 1.8,
+    "R": -4.5,
+    "N": -3.5,
+    "D": -3.5,
+    "C": 2.5,
+    "Q": -3.5,
+    "E": -3.5,
+    "G": -0.4,
+    "H": -3.2,
+    "I": 4.5,
+    "L": 3.8,
+    "K": -3.9,
+    "M": 1.9,
+    "F": 2.8,
+    "P": -1.6,
+    "S": -0.8,
+    "T": -0.7,
+    "W": -0.9,
+    "Y": -1.3,
+    "V": 4.2,
+}
+
+TM_WINDOW = 19
+TM_THRESHOLD = 1.6
+MEAN_KD_RISK_THRESHOLD = 0.3
+SP_HRUN_MIN = 7
+SP_SCAN_WINDOW = 30

factorforge_cds-3.2.1/src/factorforge/protein_risk/risk_classifier.py

@@ -0,0 +1,14 @@
+"""Protein structural-risk classification rules."""
+
+from factorforge.protein_risk.kd_scale import MEAN_KD_RISK_THRESHOLD
+
+
+def classify_risk(tm_count: int, mean_kd: float, sp_predicted: bool) -> str:
+    """Classify deterministic structural-risk indicators."""
+    if tm_count >= 3 or (
+        tm_count >= 1 and mean_kd >= MEAN_KD_RISK_THRESHOLD and not sp_predicted
+    ):
+        return "HIGH"
+    if tm_count >= 1:
+        return "MEDIUM"
+    return "LOW"

factorforge_cds-3.2.1/src/factorforge/protein_risk/sp_predict.py

@@ -0,0 +1,18 @@
+"""Deterministic N-terminal signal-peptide heuristic."""
+
+from factorforge.protein_risk.kd_scale import SP_HRUN_MIN, SP_SCAN_WINDOW
+
+HYDROPHOBIC_RESIDUES = frozenset("ILVAMFW")
+
+
+def predict_signal_peptide(protein_seq: str) -> bool:
+    """Return whether an N-terminal hydrophobic run meets the heuristic."""
+    run_length = 0
+    for residue in protein_seq.upper()[:SP_SCAN_WINDOW]:
+        if residue in HYDROPHOBIC_RESIDUES:
+            run_length += 1
+            if run_length >= SP_HRUN_MIN:
+                return True
+        else:
+            run_length = 0
+    return False

factorforge_cds-3.2.1/src/factorforge/protein_risk/tm_predict.py

@@ -0,0 +1,30 @@
+"""Deterministic transmembrane-segment heuristic."""
+
+from factorforge.protein_risk.kd_scale import KD_SCALE, TM_THRESHOLD, TM_WINDOW
+
+
+def predict_tm_segments(protein_seq: str) -> dict:
+    """Return merged hydrophobic windows and whole-protein mean KD score."""
+    if not protein_seq:
+        raise ValueError("protein_seq must not be empty")
+
+    scores = [KD_SCALE.get(residue, 0.0) for residue in protein_seq.upper()]
+    candidate_segments: list[tuple[int, int]] = []
+    for start in range(max(0, len(scores) - TM_WINDOW + 1)):
+        window_mean = sum(scores[start : start + TM_WINDOW]) / TM_WINDOW
+        if window_mean >= TM_THRESHOLD:
+            candidate_segments.append((start, start + TM_WINDOW))
+
+    merged_segments: list[tuple[int, int]] = []
+    for start, end in candidate_segments:
+        if merged_segments and start <= merged_segments[-1][1]:
+            previous_start, previous_end = merged_segments[-1]
+            merged_segments[-1] = (previous_start, max(previous_end, end))
+        else:
+            merged_segments.append((start, end))
+
+    return {
+        "tm_count": len(merged_segments),
+        "mean_kd_score": sum(scores) / len(scores),
+        "segments": merged_segments,
+    }

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge/schemas/design_package.py

@@ -17,7 +17,7 @@ Separation rationale:
     before any output is published or shared externally.
 """
 
-from typing import Any, Optional
+from typing import Any, Literal, Optional
 
 from pydantic import BaseModel, ConfigDict, Field
 
@@ -90,6 +90,16 @@ class WetLabFeedback(BaseModel):
     submissions: list[Any] = Field(default_factory=list)
 
 
+class ProteinRisk(BaseModel):
+    model_config = ConfigDict(extra="forbid")
+
+    tm_count: int = Field(ge=0)
+    mean_kd_score: float
+    signal_peptide_predicted: bool
+    risk_level: Literal["LOW", "MEDIUM", "HIGH"]
+    warnings: list[str] = Field(default_factory=list)
+
+
 class DesignPackage(BaseModel):
     model_config = ConfigDict(extra="allow")
 
@@ -103,3 +113,4 @@ class DesignPackage(BaseModel):
     validation_status: Optional[ValidationStatus] = None
     provenance: Provenance
     wet_lab_feedback: WetLabFeedback = Field(default_factory=WetLabFeedback)
+    protein_risk: Optional[ProteinRisk] = None

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge/schemas/design_package.schema.json

@@ -51,6 +51,9 @@
     },
     "claim_boundary": {
       "$ref": "#/$defs/claimBoundary"
+    },
+    "protein_risk": {
+      "$ref": "#/$defs/proteinRisk"
     }
   },
   "allOf": [
@@ -134,6 +137,39 @@
     }
   ],
   "$defs": {
+    "proteinRisk": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": [
+        "tm_count",
+        "mean_kd_score",
+        "signal_peptide_predicted",
+        "risk_level",
+        "warnings"
+      ],
+      "properties": {
+        "tm_count": {
+          "type": "integer",
+          "minimum": 0
+        },
+        "mean_kd_score": {
+          "type": "number"
+        },
+        "signal_peptide_predicted": {
+          "type": "boolean"
+        },
+        "risk_level": {
+          "type": "string",
+          "enum": ["LOW", "MEDIUM", "HIGH"]
+        },
+        "warnings": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          }
+        }
+      }
+    },
     "hostProfile": {
       "type": "object",
       "additionalProperties": false,

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge_cds.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: factorforge-cds
-Version: 3.2.0
+Version: 3.2.1
 Summary: FactorForge - open-source constraint-based CDS design engine by Eijex.
 Author-email: Eijex <eijex.lab@gmail.com>
 License-Expression: AGPL-3.0-only
@@ -21,6 +21,7 @@ Requires-Dist: click>=8.0
 Requires-Dist: pydantic>=2.0
 Provides-Extra: dev
 Requires-Dist: jsonschema>=4.0; extra == "dev"
+Requires-Dist: pandas>=2.0; extra == "dev"
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
 Requires-Dist: ruff>=0.1; extra == "dev"
@@ -39,7 +40,7 @@ Dynamic: license-file
 [![PyPI](https://img.shields.io/pypi/v/factorforge-cds.svg)](https://pypi.org/project/factorforge-cds/)
 [![CI](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml/badge.svg)](https://github.com/eijex/factorforge-cds/actions/workflows/ci.yml)
 [![codecov](https://codecov.io/gh/eijex/factorforge-cds/branch/main/graph/badge.svg)](https://codecov.io/gh/eijex/factorforge-cds)
-[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20407331.svg)](https://doi.org/10.5281/zenodo.20407331)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.20640931.svg)](https://doi.org/10.5281/zenodo.20640931)
 [![Web App](https://img.shields.io/badge/web-factorforge.eijex.com-brightgreen.svg)](https://factorforge.eijex.com)
 
 FactorForge performs profile-guided CDS design with CAI/GC metrics, PolyA-signal screening, and Golden Gate/MoClo-aware checks. Primary support: *N. benthamiana* (agroinfiltration). Experimental host context: Tobacco BY-2 (`--host by2`).
@@ -113,7 +114,7 @@ GNU Affero General Public License v3.0 — see [LICENSE](LICENSE).
 ## Get in Touch
 
 - **Docs** — [eijex.github.io/factorforge-cds](https://eijex.github.io/factorforge-cds/)
-- **Wet-lab Results** — Public-safe validation summaries are welcome. Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
+- **Wet-lab Results** — Public-safe validation summaries are welcome. [Share Wet-lab Results (Form)](https://docs.google.com/forms/d/e/1FAIpQLSeSx-wYvF6YwHhSPdLMl-L44frCugdm25X_eDz50OaqTD66qA/viewform) or [Share Wet-lab Results (GitHub)](https://github.com/eijex/factorforge-cds/issues/new?template=wet_lab_result.yml) (public-safe summaries only). Do not submit raw sequences, confidential construct details, internal batch IDs, patient data, private contact information, exact process parameters, or confidential partner/customer data. See [VALIDATION.md](VALIDATION.md) before submitting.
 - **GitHub Issues** — bugs, features: [github.com/eijex/factorforge-cds/issues](https://github.com/eijex/factorforge-cds/issues)
 - **Email** — eijex.lab@gmail.com
 - **FactorForge** — [factorforge.eijex.com](https://factorforge.eijex.com)

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge_cds.egg-info/SOURCES.txt

@@ -39,6 +39,12 @@ src/factorforge/engines/profile/rules/rule_engine.py
 src/factorforge/io/__init__.py
 src/factorforge/io/fasta.py
 src/factorforge/io/validation.py
+src/factorforge/protein_risk/__init__.py
+src/factorforge/protein_risk/annotate.py
+src/factorforge/protein_risk/kd_scale.py
+src/factorforge/protein_risk/risk_classifier.py
+src/factorforge/protein_risk/sp_predict.py
+src/factorforge/protein_risk/tm_predict.py
 src/factorforge/registry/__init__.py
 src/factorforge/registry/registry_loader.py
 src/factorforge/schemas/__init__.py
@@ -70,6 +76,7 @@ tests/test_database.py
 tests/test_design_package_schema.py
 tests/test_design_package_semantics.py
 tests/test_design_package_serialization.py
+tests/test_docs_consistency.py
 tests/test_fasta_io.py
 tests/test_gc_content.py
 tests/test_host_profile_metadata.py
@@ -78,7 +85,9 @@ tests/test_legacy_cli.py
 tests/test_no_raw_sequence_logging.py
 tests/test_openbio_missing_metric_contract.py
 tests/test_parameter_registry.py
+tests/test_protein_risk.py
 tests/test_registry_production_sync.py
 tests/test_restriction_sites.py
 tests/test_sequence_validator.py
-tests/test_translation_integrity.py
+tests/test_translation_integrity.py
+tests/test_worked_example.py

{factorforge_cds-3.2.0 → factorforge_cds-3.2.1}/src/factorforge_cds.egg-info/requires.txt

@@ -5,6 +5,7 @@ pydantic>=2.0
 
 [dev]
 jsonschema>=4.0
+pandas>=2.0
 pytest>=7.0
 pytest-cov>=4.0
 ruff>=0.1

factorforge_cds-3.2.1/tests/test_benchmark_scoring.py

@@ -0,0 +1,98 @@
+import dataclasses
+import pandas as pd
+import pytest
+from benchmarks.config import load_benchmark_config
+from benchmarks.scoring import score_cds, canonical_multi_constraint_pass
+
+CFG = load_benchmark_config()
+
+
+def test_score_cds_full_schema():
+    row = score_cds("random_synonymous", "baseline", "p1", "MKT", "ATGAAAACC", CFG, runtime_seconds=0.01)
+    expected = {"method","method_type","sequence_id","aa_identity","internal_stop_count",
+                "invalid_codon_count","length_multiple_of_three","cai","gc_percent",
+                "gc_in_target_range","forbidden_type_iis_site_count","biological_pass",
+                "assembly_pass","multi_constraint_pass","runtime_seconds"}
+    assert expected.issubset(row.keys())
+    assert row["biological_pass"] is True
+    assert row["aa_identity"] == 1.0
+
+
+# --- scoring_contract v1.1 semantic invariant tests ---
+
+def test_multi_constraint_pass_requires_gc_in_range():
+    """multi_constraint_pass=True implies gc_in_target_range=True (scoring_contract v1.1)."""
+    # Use a very narrow GC window that forces gc_in_target_range=False for a typical CDS
+    cfg_narrow = dataclasses.replace(CFG, gc_min=99.0, gc_max=100.0)
+    row = score_cds("greedy_cai", "baseline", "t1", "MKT", "ATGAAAACC", cfg_narrow, runtime_seconds=0.01)
+    assert row["gc_in_target_range"] is False
+    assert row["multi_constraint_pass"] is False, (
+        "multi_constraint_pass must be False when gc_in_target_range is False (scoring_contract v1.1)"
+    )
+
+
+def test_multi_constraint_pass_true_only_when_all_primitives_pass():
+    """multi_constraint_pass=True only when biological_pass AND assembly_pass AND gc_in_target_range."""
+    # MKT codon ATGAAAACC: GC = (0+0+0+0+0+1+0+1+1)/9 = 3/9 ~ 33%
+    # Default gc_min=55, gc_max=65 → gc_in_target_range=False → multi_constraint_pass=False
+    row = score_cds("greedy_cai", "baseline", "t2", "MKT", "ATGAAAACC", CFG, runtime_seconds=0.01)
+    if not row["gc_in_target_range"]:
+        assert row["multi_constraint_pass"] is False
+    if not row["assembly_pass"]:
+        assert row["multi_constraint_pass"] is False
+
+
+def test_canonical_helper_gc_fail_implies_false():
+    """canonical_multi_constraint_pass: gc_in_target_range=False → result=False."""
+    df = pd.DataFrame([{
+        "biological_pass": True,
+        "assembly_pass": True,
+        "gc_in_target_range": False,
+    }])
+    result = canonical_multi_constraint_pass(df, gc_min=55.0, gc_max=65.0)
+    assert bool(result.iloc[0]) is False
+
+
+def test_canonical_helper_all_pass_implies_true():
+    """canonical_multi_constraint_pass: all primitives True → result=True."""
+    df = pd.DataFrame([{
+        "biological_pass": True,
+        "assembly_pass": True,
+        "gc_in_target_range": True,
+    }])
+    result = canonical_multi_constraint_pass(df, gc_min=55.0, gc_max=65.0)
+    assert bool(result.iloc[0]) is True
+
+
+def test_canonical_helper_assembly_fail_implies_false():
+    """canonical_multi_constraint_pass: assembly_pass=False → result=False."""
+    df = pd.DataFrame([{
+        "biological_pass": True,
+        "assembly_pass": False,
+        "gc_in_target_range": True,
+    }])
+    result = canonical_multi_constraint_pass(df, gc_min=55.0, gc_max=65.0)
+    assert bool(result.iloc[0]) is False
+
+
+def test_canonical_helper_biological_fail_implies_false():
+    """canonical_multi_constraint_pass: biological_pass=False → result=False."""
+    df = pd.DataFrame([{
+        "biological_pass": False,
+        "assembly_pass": True,
+        "gc_in_target_range": True,
+    }])
+    result = canonical_multi_constraint_pass(df, gc_min=55.0, gc_max=65.0)
+    assert bool(result.iloc[0]) is False
+
+
+def test_canonical_helper_nan_treated_as_false():
+    """canonical_multi_constraint_pass: NaN in any primitive → result=False."""
+    import numpy as np
+    df = pd.DataFrame([{
+        "biological_pass": True,
+        "assembly_pass": True,
+        "gc_in_target_range": float("nan"),
+    }])
+    result = canonical_multi_constraint_pass(df, gc_min=55.0, gc_max=65.0)
+    assert bool(result.iloc[0]) is False

factorforge_cds-3.2.1/tests/test_docs_consistency.py

@@ -0,0 +1,143 @@
+"""Docs-as-Code consistency guardrails (job 108).
+
+Checks that key documentation files stay in sync with their source-of-truth
+artifacts. Fails CI if claim wording, ablation layer definitions, or
+reproducibility anchors drift.
+"""
+from __future__ import annotations
+import hashlib
+import json
+from pathlib import Path
+
+import pytest
+import yaml
+
+ROOT = Path(__file__).resolve().parents[1]
+BENCH_DIR = ROOT / "reproducibility" / "benchmark_v0.5.1"
+ABLATION_SPEC = ROOT / "benchmarks" / "ablation" / "ablation_spec.yaml"
+BENCHMARK_README = BENCH_DIR / "README.md"
+MANIFEST_PATH = BENCH_DIR / "MANIFEST.json"
+FROZEN_SUMMARY = BENCH_DIR / "data" / "benchmark_summary.frozen.json"
+
+
+# ---------------------------------------------------------------------------
+# Reproducibility anchor files
+# ---------------------------------------------------------------------------
+
+def test_manifest_json_exists_and_valid():
+    assert MANIFEST_PATH.exists(), f"Missing: {MANIFEST_PATH}"
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    for key in ("factorforge_version", "git_commit", "scoring_contract", "inputs", "outputs"):
+        assert key in data, f"MANIFEST.json missing key: {key}"
+
+
+def test_benchmark_summary_frozen_exists():
+    assert FROZEN_SUMMARY.exists(), f"Missing: {FROZEN_SUMMARY}"
+
+
+def test_manifest_scoring_contract_v1_1():
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    assert data["scoring_contract"] == "v1.1", (
+        f"Expected scoring_contract='v1.1', got {data['scoring_contract']!r}"
+    )
+
+
+def test_manifest_inputs_sha256_non_empty():
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    for name, entry in data["inputs"].items():
+        sha = entry.get("sha256", "")
+        assert sha and len(sha) == 64, (
+            f"MANIFEST.json inputs[{name!r}].sha256 is missing or invalid"
+        )
+
+
+def test_manifest_input_sha256_matches_files():
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    for name, entry in data["inputs"].items():
+        path = ROOT / entry["path"]
+        actual = hashlib.sha256(path.read_bytes()).hexdigest()
+        assert actual == entry["sha256"], f"MANIFEST.json hash drift for {name}: {path}"
+
+
+def test_manifest_commands_use_reproducible_entrypoints():
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    commands = "\n".join(data["commands"])
+    assert "python scripts/benchmark.py" not in commands
+    assert "python -m benchmarks.run_benchmark" in commands
+    assert "--dataset nbenthamiana_full --mode formal --seed 320" in commands
+    assert "python -m benchmarks.ablation.run_ablation" in commands
+
+
+def test_manifest_separates_software_and_benchmark_dois():
+    data = json.loads(MANIFEST_PATH.read_text(encoding="utf-8"))
+    assert data["archives"]["software_release"]["doi"] == "10.5281/zenodo.20640931"
+    assert data["archives"]["corrected_benchmark_dataset"]["doi"] == "10.5281/zenodo.20676276"
+
+
+# ---------------------------------------------------------------------------
+# Ablation spec ↔ benchmark README consistency
+# ---------------------------------------------------------------------------
+
+@pytest.fixture(scope="module")
+def ablation_spec():
+    return yaml.safe_load(ABLATION_SPEC.read_text(encoding="utf-8"))
+
+
+@pytest.fixture(scope="module")
+def benchmark_readme():
+    return BENCHMARK_README.read_text(encoding="utf-8")
+
+
+def test_ablation_spec_layers_in_benchmark_readme(ablation_spec, benchmark_readme):
+    """Every layer key (L0–L5) from ablation_spec.yaml must appear in the benchmark README."""
+    layers = ablation_spec.get("layers", {})
+    assert layers, "ablation_spec.yaml has no 'layers' section"
+    for level in layers:
+        assert level in benchmark_readme, (
+            f"ablation_spec layer '{level}' not documented in benchmark README"
+        )
+
+
+def test_ablation_spec_layer_names_in_benchmark_readme(ablation_spec, benchmark_readme):
+    """Every layer 'name' value from ablation_spec.yaml must appear in the benchmark README."""
+    layers = ablation_spec.get("layers", {})
+    for level, cfg in layers.items():
+        name = cfg.get("name", "")
+        assert name and name in benchmark_readme, (
+            f"Layer {level} name '{name}' not found in benchmark README"
+        )
+
+
+# ---------------------------------------------------------------------------
+# Evidence boundary / claim wording
+# ---------------------------------------------------------------------------
+
+def test_main_readme_evidence_boundary():
+    """Main README must contain in-silico or computational disclaimer."""
+    readme = (ROOT / "README.md").read_text(encoding="utf-8")
+    assert "in-silico" in readme or "computational" in readme, (
+        "Main README missing evidence boundary disclaimer (in-silico / computational)"
+    )
+
+
+def test_benchmark_readme_evidence_boundary(benchmark_readme):
+    """Benchmark README must contain in-silico or computational disclaimer."""
+    assert "in-silico" in benchmark_readme or "computational" in benchmark_readme, (
+        "Benchmark README missing evidence boundary disclaimer"
+    )
+
+
+def test_benchmark_readme_no_wet_lab_claim(benchmark_readme):
+    """Benchmark README must not contain unsupported wet-lab outcome claims."""
+    forbidden = [
+        "predicts expression",
+        "predicts yield",
+        "wet-lab success",
+        "guarantees cloning",
+        "guarantees synthesis",
+        "predicts synthesis success",
+    ]
+    for phrase in forbidden:
+        assert phrase.lower() not in benchmark_readme.lower(), (
+            f"Benchmark README contains forbidden claim phrase: {phrase!r}"
+        )