Merge pull request #519 from VariantEffect/feature/bencap/517/additional-brnich-style-calibrations

Additional Brnich Style Calibrations

Author: bencap

alembic/manual_migrations/migrate_consolidated_score_ranges_and_calibrations.py

@@ -9,7 +9,7 @@
 from sqlalchemy.orm import Session
 
 from mavedb.models.score_set import ScoreSet
-from mavedb.view_models.score_range import ScoreSetRangesCreate, InvestigatorScoreRangesCreate, PillarProjectScoreRangesCreate, PillarProjectScoreRangeCreate
+from mavedb.view_models.score_range import ScoreSetRangesCreate, InvestigatorScoreRangesCreate, ZeibergCalibrationScoreRangesCreate, ZeibergCalibrationScoreRangeCreate
 
 
 from mavedb.db.session import SessionLocal
@@ -47,11 +47,11 @@ def do_migration(db: Session):
             investigator_ranges = None
 
         if score_set.score_calibrations is not None:
-            thresholds = score_set.score_calibrations.get("pillar_project", {}).get("thresholds", [])
-            evidence_strengths = score_set.score_calibrations.get("pillar_project", {}).get("evidence_strengths", [])
-            positive_likelihood_ratios = score_set.score_calibrations.get("pillar_project", {}).get("positive_likelihood_ratios", [])
-            prior_probability_pathogenicity = score_set.score_calibrations.get("pillar_project", {}).get("prior_probability_pathogenicity", None)
-            parameter_sets = score_set.score_calibrations.get("pillar_project", {}).get("parameter_sets", [])
+            thresholds = score_set.score_calibrations.get("zeiberg_calibration", {}).get("thresholds", [])
+            evidence_strengths = score_set.score_calibrations.get("zeiberg_calibration", {}).get("evidence_strengths", [])
+            positive_likelihood_ratios = score_set.score_calibrations.get("zeiberg_calibration", {}).get("positive_likelihood_ratios", [])
+            prior_probability_pathogenicity = score_set.score_calibrations.get("zeiberg_calibration", {}).get("prior_probability_pathogenicity", None)
+            parameter_sets = score_set.score_calibrations.get("zeiberg_calibration", {}).get("parameter_sets", [])
 
             ranges = []
             boundary_direction = -1  # Start with a negative sign to indicate the first range has the lower boundary appearing prior to the threshold
@@ -60,7 +60,7 @@ def do_migration(db: Session):
 
                 if idx == 0:
                     calculated_range = (None, threshold)
-                    ranges.append(PillarProjectScoreRangeCreate(
+                    ranges.append(ZeibergCalibrationScoreRangeCreate(
                         range=(None, threshold),
                         classification="normal" if evidence_strength < 0 else "abnormal",
                         label=str(evidence_strength),
@@ -71,7 +71,7 @@ def do_migration(db: Session):
                     ))
                 elif idx == len(thresholds) - 1:
                     calculated_range = (threshold, None)
-                    ranges.append(PillarProjectScoreRangeCreate(
+                    ranges.append(ZeibergCalibrationScoreRangeCreate(
                         range=(threshold, None),
                         classification="normal" if evidence_strength < 0 else "abnormal",
                         evidence_strength=evidence_strength,
@@ -86,7 +86,7 @@ def do_migration(db: Session):
                     else:
                         calculated_range = (threshold, thresholds[idx + 1])
 
-                    ranges.append(PillarProjectScoreRangeCreate(
+                    ranges.append(ZeibergCalibrationScoreRangeCreate(
                         range=calculated_range,
                         classification="normal" if evidence_strength < 0 else "abnormal",
                         label=str(evidence_strength),
@@ -100,17 +100,17 @@ def do_migration(db: Session):
                 if idx != len(evidence_strengths) - 1 and (evidence_strengths[idx + 1] * evidence_strength < 0):
                     boundary_direction = -boundary_direction
 
-                pillar_project_ranges = PillarProjectScoreRangesCreate(
+                zeiberg_calibration_ranges = ZeibergCalibrationScoreRangesCreate(
                     prior_probability_pathogenicity=prior_probability_pathogenicity,
                     parameter_sets=parameter_sets,
                     ranges=ranges,
                 )
         else:
-            pillar_project_ranges = None
+            zeiberg_calibration_ranges = None
 
         score_set.score_ranges = ScoreSetRangesCreate(
             investigator_provided=investigator_ranges if investigator_ranges else None,
-            pillar_project=pillar_project_ranges if pillar_project_ranges else None,
+            zeiberg_calibration=zeiberg_calibration_ranges if zeiberg_calibration_ranges else None,
         ).model_dump()
         db.add(score_set)
 

src/mavedb/lib/annotation/agent.py

@@ -58,14 +58,14 @@ def mavedb_user_agent(user: User) -> Agent:
 
 
 # XXX: Ideally, this becomes versioned software.
-def pillar_project_calibration_agent() -> Agent:
+def zeiberg_calibration_agent() -> Agent:
     """
     Create a [VA Agent](https://va-ga4gh.readthedocs.io/en/latest/core-information-model/entities/agent.html)
-    object for the pillar project calibration software.
+    object for the Zeiberg calibration software.
     """
     return Agent(
-        name="Pillar Project Variant Calibrator",
+        name="Zeiberg Variant Calibrator",
         agentType="Software",
         # XXX - version?
-        description="Pillar project variant calibrator, see https://github.com/Dzeiberg/mave_calibration",
+        description="Zeiberg variant calibrator, see https://github.com/Dzeiberg/mave_calibration",
     )

src/mavedb/lib/annotation/classification.py

@@ -6,7 +6,7 @@
 from ga4gh.va_spec.base.enums import StrengthOfEvidenceProvided
 
 from mavedb.models.mapped_variant import MappedVariant
-from mavedb.lib.annotation.constants import PILLAR_PROJECT_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP
+from mavedb.lib.annotation.constants import ZEIBERG_CALIBRATION_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP
 from mavedb.lib.validation.utilities import inf_or_float
 from mavedb.view_models.score_range import ScoreSetRanges
 
@@ -60,7 +60,7 @@ def functional_classification_of_variant(
     return ExperimentalVariantFunctionalImpactClassification.INDETERMINATE
 
 
-def pillar_project_clinical_classification_of_variant(
+def zeiberg_calibration_clinical_classification_of_variant(
     mapped_variant: MappedVariant,
 ) -> tuple[VariantPathogenicityEvidenceLine.Criterion, Optional[StrengthOfEvidenceProvided]]:
     if mapped_variant.variant.score_set.score_ranges is None:
@@ -69,7 +69,7 @@ def pillar_project_clinical_classification_of_variant(
             " Unable to classify clinical impact."
         )
 
-    score_ranges = ScoreSetRanges(**mapped_variant.variant.score_set.score_ranges).pillar_project
+    score_ranges = ScoreSetRanges(**mapped_variant.variant.score_set.score_ranges).zeiberg_calibration
 
     if not score_ranges:
         raise ValueError(
@@ -88,6 +88,6 @@ def pillar_project_clinical_classification_of_variant(
     for range in score_ranges.ranges:
         lower_bound, upper_bound = inf_or_float(range.range[0], lower=True), inf_or_float(range.range[1], lower=False)
         if functional_score > lower_bound and functional_score <= upper_bound:
-            return PILLAR_PROJECT_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP[range.evidence_strength]
+            return ZEIBERG_CALIBRATION_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP[range.evidence_strength]
 
-    return PILLAR_PROJECT_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP[0]
+    return ZEIBERG_CALIBRATION_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP[0]

src/mavedb/lib/annotation/constants.py

@@ -4,7 +4,7 @@
 GENERIC_DISEASE_MEDGEN_CODE = "C0012634"
 MEDGEN_SYSTEM = "https://www.ncbi.nlm.nih.gov/medgen/"
 
-PILLAR_PROJECT_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP = {
+ZEIBERG_CALIBRATION_CALIBRATION_STRENGTH_OF_EVIDENCE_MAP = {
     # No evidence
     0: (VariantPathogenicityEvidenceLine.Criterion.PS3, None),
     # Supporting evidence
@@ -31,4 +31,4 @@
 
 # TODO#493
 FUNCTIONAL_RANGES = ["investigator_provided"]
-CLINICAL_RANGES = ["pillar_project"]
+CLINICAL_RANGES = ["zeiberg_calibration"]

src/mavedb/lib/annotation/contribution.py

@@ -10,7 +10,7 @@
     mavedb_api_agent,
     mavedb_vrs_agent,
     mavedb_user_agent,
-    pillar_project_calibration_agent,
+    zeiberg_calibration_agent,
 )
 from mavedb.lib.types.annotation import ResourceWithCreationModificationDates
 
@@ -46,15 +46,15 @@ def mavedb_vrs_contribution(mapped_variant: MappedVariant) -> Contribution:
     )
 
 
-def pillar_project_calibration_contribution() -> Contribution:
+def zeiberg_calibration_contribution() -> Contribution:
     """
     Create a [VA Contribution](https://va-ga4gh.readthedocs.io/en/latest/core-information-model/entities/activities/contribution.html#contribution)
     object for a sofware agent which performs calibrations on an arbitrary data set.
     """
     return Contribution(
-        name="MaveDB Pillar Project Calibration",
-        description="Contribution from the MaveDB Pillar Project Calibration software",
-        contributor=pillar_project_calibration_agent(),
+        name="Zeiberg Calibration",
+        description="Contribution from the Zeiberg Calibration software",
+        contributor=zeiberg_calibration_agent(),
         activityType="variant specific calibration software",
     )
 

src/mavedb/lib/annotation/evidence_line.py

@@ -12,17 +12,17 @@
     VariantPathogenicityProposition,
 )
 
-from mavedb.lib.annotation.classification import pillar_project_clinical_classification_of_variant
+from mavedb.lib.annotation.classification import zeiberg_calibration_clinical_classification_of_variant
 from mavedb.lib.annotation.contribution import (
     mavedb_api_contribution,
     mavedb_vrs_contribution,
-    pillar_project_calibration_contribution,
+    zeiberg_calibration_contribution,
     mavedb_creator_contribution,
     mavedb_modifier_contribution,
 )
 from mavedb.lib.annotation.document import score_set_to_document
 from mavedb.lib.annotation.method import (
-    pillar_project_calibration_method,
+    zeiberg_calibration_method,
     publication_identifiers_to_method,
 )
 from mavedb.models.mapped_variant import MappedVariant
@@ -33,7 +33,7 @@ def acmg_evidence_line(
     proposition: VariantPathogenicityProposition,
     evidence: list[Union[StudyResult, EvidenceLineType, StatementType, iriReference]],
 ) -> Optional[VariantPathogenicityEvidenceLine]:
-    evidence_outcome, evidence_strength = pillar_project_clinical_classification_of_variant(mapped_variant)
+    evidence_outcome, evidence_strength = zeiberg_calibration_clinical_classification_of_variant(mapped_variant)
 
     if not evidence_strength:
         evidence_outcome_code = f"{evidence_outcome.value}_not_met"
@@ -59,7 +59,7 @@ def acmg_evidence_line(
 
     return VariantPathogenicityEvidenceLine(
         description=f"Pathogenicity evidence line {mapped_variant.variant.urn}.",
-        specifiedBy=pillar_project_calibration_method(evidence_outcome),
+        specifiedBy=zeiberg_calibration_method(evidence_outcome),
         evidenceOutcome={
             "primaryCoding": Coding(
                 code=evidence_outcome_code,
@@ -72,7 +72,7 @@ def acmg_evidence_line(
         contributions=[
             mavedb_api_contribution(),
             mavedb_vrs_contribution(mapped_variant),
-            pillar_project_calibration_contribution(),
+            zeiberg_calibration_contribution(),
             mavedb_creator_contribution(mapped_variant.variant, mapped_variant.variant.score_set.created_by),
             mavedb_modifier_contribution(mapped_variant.variant, mapped_variant.variant.score_set.modified_by),
         ],

src/mavedb/lib/annotation/method.py

@@ -86,22 +86,22 @@ def mavedb_vrs_as_method() -> Method:
     return Method(name="Software version", reportedIn=mavedb_vrs_releases_as_iri())
 
 
-def pillar_project_calibrations_as_iri() -> IRI:
+def zeiberg_calibrations_as_iri() -> IRI:
     """
     Create an IRI as described in <https://datatracker.ietf.org/doc/html/rfc3986#section-4.1> for the software used to generate pillar project calibrations. Within
     the context of VA-Spec, this IRI can be used interchangeably with an equivalent method object for brevity.
     """
     return IRI("https://github.com/Dzeiberg/mave_calibration")
 
 
-def pillar_project_calibration_method(method: Optional[VariantPathogenicityEvidenceLine.Criterion]) -> Method:
+def zeiberg_calibration_method(method: Optional[VariantPathogenicityEvidenceLine.Criterion]) -> Method:
     """
     Generate a [VA Method](https://va-ga4gh.readthedocs.io/en/latest/core-information-model/entities/information-entities/method.html#method)
     object for the pillar project calibration software distribution.
     """
     return Method(
         name="Software version",
-        reportedIn=pillar_project_calibrations_as_iri(),
+        reportedIn=zeiberg_calibrations_as_iri(),
         methodType=method.value if method else None,
     )
 

src/mavedb/scripts/load_pp_style_calibration.py

@@ -9,8 +9,8 @@
 from mavedb.scripts.environment import with_database_session
 from mavedb.models.score_set import ScoreSet
 from mavedb.view_models.score_range import (
-    PillarProjectScoreRangeCreate,
-    PillarProjectScoreRangesCreate,
+    ZeibergCalibrationScoreRangeCreate,
+    ZeibergCalibrationScoreRangesCreate,
     ScoreSetRangesCreate,
 )
 
@@ -47,7 +47,9 @@ def _collapse_duplicate_thresholds(m: dict[int, Optional[float]], comparator: Ca
     return collapsed
 
 
-def build_pathogenic_ranges(thresholds: List[Optional[float]], inverted: bool) -> List[PillarProjectScoreRangeCreate]:
+def build_pathogenic_ranges(
+    thresholds: List[Optional[float]], inverted: bool
+) -> List[ZeibergCalibrationScoreRangeCreate]:
     raw_mapping = {
         strength: thresholds[idx]
         for idx, strength in enumerate(PATH_STRENGTHS)
@@ -61,7 +63,7 @@ def build_pathogenic_ranges(thresholds: List[Optional[float]], inverted: bool) -
     available = [s for s in PATH_STRENGTHS if s in mapping]
     ordering = available[::-1] if not inverted else available
 
-    ranges: List[PillarProjectScoreRangeCreate] = []
+    ranges: List[ZeibergCalibrationScoreRangeCreate] = []
     for i, s in enumerate(ordering):
         lower: Optional[float]
         upper: Optional[float]
@@ -74,7 +76,7 @@ def build_pathogenic_ranges(thresholds: List[Optional[float]], inverted: bool) -
             upper = mapping[s]
 
         ranges.append(
-            PillarProjectScoreRangeCreate(
+            ZeibergCalibrationScoreRangeCreate(
                 label=str(s),
                 classification="abnormal",
                 evidence_strength=s,
@@ -87,7 +89,7 @@ def build_pathogenic_ranges(thresholds: List[Optional[float]], inverted: bool) -
     return ranges
 
 
-def build_benign_ranges(thresholds: List[Optional[float]], inverted: bool) -> List[PillarProjectScoreRangeCreate]:
+def build_benign_ranges(thresholds: List[Optional[float]], inverted: bool) -> List[ZeibergCalibrationScoreRangeCreate]:
     raw_mapping = {
         strength: thresholds[idx]
         for idx, strength in enumerate(BENIGN_STRENGTHS)
@@ -101,7 +103,7 @@ def build_benign_ranges(thresholds: List[Optional[float]], inverted: bool) -> Li
     available = [s for s in BENIGN_STRENGTHS if s in mapping]
     ordering = available[::-1] if inverted else available
 
-    ranges: List[PillarProjectScoreRangeCreate] = []
+    ranges: List[ZeibergCalibrationScoreRangeCreate] = []
     for i, s in enumerate(ordering):
         lower: Optional[float]
         upper: Optional[float]
@@ -114,7 +116,7 @@ def build_benign_ranges(thresholds: List[Optional[float]], inverted: bool) -> Li
             upper = mapping[s]
 
         ranges.append(
-            PillarProjectScoreRangeCreate(
+            ZeibergCalibrationScoreRangeCreate(
                 label=str(s),
                 classification="normal",
                 evidence_strength=s,
@@ -133,12 +135,12 @@ def build_benign_ranges(thresholds: List[Optional[float]], inverted: bool) -> Li
 @click.argument("score_set_urn", type=str)
 @click.option("--overwrite", is_flag=True, default=False, help="Overwrite existing score_ranges if present.")
 def main(db: Session, json_path: str, score_set_urn: str, overwrite: bool) -> None:
-    """Load pillar project calibration JSON into a score set's pillar_project score ranges."""
+    """Load pillar project calibration JSON into a score set's zeiberg_calibration score ranges."""
     score_set: Optional[ScoreSet] = db.query(ScoreSet).filter(ScoreSet.urn == score_set_urn).one_or_none()
     if not score_set:
         raise click.ClickException(f"Score set with URN {score_set_urn} not found")
 
-    if score_set.score_ranges and score_set.score_ranges["pillar_project"] and not overwrite:
+    if score_set.score_ranges and score_set.score_ranges["zeiberg_calibration"] and not overwrite:
         raise click.ClickException(
             "pillar project score ranges already present for this score set. Use --overwrite to replace them."
         )
@@ -162,7 +164,7 @@ def main(db: Session, json_path: str, score_set_urn: str, overwrite: bool) -> No
     if not path_ranges and not benign_ranges:
         raise click.ClickException("No valid thresholds found to build ranges.")
 
-    existing_score_ranges.pillar_project = PillarProjectScoreRangesCreate(ranges=path_ranges + benign_ranges)
+    existing_score_ranges.zeiberg_calibration = ZeibergCalibrationScoreRangesCreate(ranges=path_ranges + benign_ranges)
     score_set.score_ranges = existing_score_ranges.model_dump(exclude_none=True)
 
     db.add(score_set)