Merge pull request #708 from opentargets/dev

chore: trigger release process

.github/workflows/release.yaml

@@ -34,14 +34,15 @@ jobs:
           fetch-depth: 0
           repository: opentargets/gentropy
           token: ${{ secrets.GITHUB_TOKEN }}
+          persist-credentials: false
 
       - name: Python Semantic Release
         id: semrelease
         # v9.6.0 is required due to the python v3.12 in the newer version of semantic release action which
         # breaks the poetry build command.
         uses: python-semantic-release/[email protected]
         with:
-          github_token: ${{ secrets.GITHUB_TOKEN }}
+          github_token: ${{ steps.trigger-token.outputs.token }}
 
       - name: Publish package to GitHub Release
         uses: python-semantic-release/upload-to-gh-release@main
@@ -51,7 +52,7 @@ jobs:
           # requires using GH_APP to authenitcate, otherwise push authorised with
           # the GITHUB_TOKEN does not trigger the tag artifact workflow.
           # see https://github.com/actions/create-github-app-token
-          github_token: ${{ steps.trigger-token.outputs.token }}
+          github_token: ${{ secrets.GITHUB_TOKEN }}
           tag: ${{ steps.semrelease.outputs.tag }}
 
       - name: Store the distribution packages

pyproject.toml

@@ -37,7 +37,7 @@ google-cloud-secret-manager = "^2.20.0"
 
 [tool.poetry.dev-dependencies]
 pre-commit = "^3.7.0"
-mypy = "^1.10"
+mypy = "^1.11"
 pep8-naming = "^0.14.1"
 interrogate = "^1.7.0"
 isort = "^5.13.2"
@@ -75,7 +75,7 @@ apache-airflow = "^2.8.0"
 apache-airflow-providers-google = "^10.13.1"
 pydoclint = ">=0.3.8,<0.6.0"
 prettier = "^0.0.7"
-deptry = ">=0.12,<0.17"
+deptry = ">=0.12,<0.18"
 yamllint = "^1.33.0"
 
 [tool.semantic_release]

src/gentropy/config.py

@@ -427,9 +427,9 @@ class FinemapperConfig(StepConfig):
             "start_hail": True,
         }
     )
-    study_locus_to_finemap: str = MISSING
     study_index_path: str = MISSING
-    output_path: str = MISSING
+    study_locus_manifest_path: str = MISSING
+    study_locus_index: int = MISSING
     max_causal_snps: int = MISSING
     primary_signal_pval_threshold: float = MISSING
     secondary_signal_pval_threshold: float = MISSING

src/gentropy/susie_finemapper.py

@@ -44,9 +44,9 @@ class SusieFineMapperStep:
     def __init__(
         self,
         session: Session,
-        study_locus_to_finemap: str,
         study_index_path: str,
-        output_path: str,
+        study_locus_manifest_path: str,
+        study_locus_index: int,
         max_causal_snps: int = 10,
         primary_signal_pval_threshold: float = 1,
         secondary_signal_pval_threshold: float = 1,
@@ -65,9 +65,9 @@ def __init__(
 
         Args:
             session (Session): Spark session
-            study_locus_to_finemap (str): path to the study locus to fine-map
             study_index_path (str): path to the study index
-            output_path (str): path to the output
+            study_locus_manifest_path (str): Path to the CSV manifest containing all study locus input and output locations. Should contain two columns: study_locus_input and study_locus_output
+            study_locus_index (int): Index (0-based) of the locus in the manifest to process in this call
             max_causal_snps (int): Maximum number of causal variants in locus, default is 10
             primary_signal_pval_threshold (float): p-value threshold for the lead variant from the primary signal (credibleSetIndex==1), default is 5e-8
             secondary_signal_pval_threshold (float): p-value threshold for the lead variant from the secondary signals, default is 1e-7
@@ -82,9 +82,15 @@ def __init__(
             imputed_r2_threshold (float): imputed R2 threshold, default is 0.9
             ld_score_threshold (float): LD score threshold ofr imputation, default is 5
         """
+        # Read locus manifest.
+        study_locus_manifest = pd.read_csv(study_locus_manifest_path)
+        row = study_locus_manifest.loc[study_locus_index]
+        study_locus_input = row["study_locus_input"]
+        study_locus_output = row["study_locus_output"]
+
         # Read studyLocus
         study_locus = (
-            StudyLocus.from_parquet(session, study_locus_to_finemap)
+            StudyLocus.from_parquet(session, study_locus_input)
             .df.withColumn(
                 "studyLocusId", StudyLocus.assign_study_locus_id("studyId", "variantId")
             )
@@ -115,11 +121,11 @@ def __init__(
         if result_logging is not None:
             # Write result
             result_logging["study_locus"].df.write.mode(session.write_mode).parquet(
-                output_path
+                study_locus_output
             )
             # Write log
             result_logging["log"].to_parquet(
-                output_path + ".log",
+                study_locus_output + ".log",
                 engine="pyarrow",
                 index=False,
             )