fix(ukb_ppp_study_index): update column name to match schema

opentargets · Oct 1, 2024 · ce9ae03 · ce9ae03
1 parent c3b8c2c
commit ce9ae03
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 19 deletions.
diff --git a/notebooks/Release_QC_metrics.ipynb b/notebooks/Release_QC_metrics.ipynb
@@ -419,7 +419,7 @@
     "# Number of studies\n",
     "eqtl_index=session.spark.read.parquet(eqtl_index_path, recursiveFileLookup=True)\n",
     "# Number of tissues, list of tissues\n",
-    "#eqtl_index.select(f.col(\"tissueFromSourceId\")).distinct().show(truncate=False)\n",
+    "#eqtl_index.select(f.col(\"biosampleFromSourceId\")).distinct().show(truncate=False)\n",
     "\n",
     "# Credible_set. Please use Daniels’ notebook as a reference. For each subfolder:\n",
     "# eqtl catalog susie:\n",

diff --git a/src/gentropy/datasource/ukb_ppp_eur/study_index.py b/src/gentropy/datasource/ukb_ppp_eur/study_index.py
@@ -1,4 +1,5 @@
 """Study Index for Finngen data source."""
+
 from __future__ import annotations
 
 import pyspark.sql.functions as f
@@ -29,9 +30,7 @@ def from_source(
         """
         # In order to populate the nSamples column, we need to peek inside the summary stats dataframe.
         num_of_samples = (
-            spark
-            .read
-            .parquet(raw_summary_stats_path)
+            spark.read.parquet(raw_summary_stats_path)
             .filter(f.col("chromosome") == "22")
             .groupBy("studyId")
             .agg(f.first("N").cast("integer").alias("nSamples"))
@@ -45,29 +44,25 @@ def from_source(
                 f.lit("UKB_PPP_EUR").alias("projectId"),
                 f.col("_gentropy_study_id").alias("studyId"),
                 f.col("UKBPPP_ProteinID").alias("traitFromSource"),
-                f.lit("UBERON_0001969").alias("tissueFromSourceId"),
+                f.lit("UBERON_0001969").alias("biosampleFromSourceId"),
                 f.col("ensembl_id").alias("geneId"),
                 f.lit(True).alias("hasSumstats"),
                 f.col("_gentropy_summary_stats_link").alias("summarystatsLocation"),
             )
             .join(num_of_samples, "studyId", "inner")
         )
         # Add population structure.
-        study_index_df = (
-            study_index_df
-            .withColumn(
-                "discoverySamples",
-                f.array(
-                    f.struct(
-                        f.col("nSamples").cast("integer").alias("sampleSize"),
-                        f.lit("European").alias("ancestry"),
-                    )
+        study_index_df = study_index_df.withColumn(
+            "discoverySamples",
+            f.array(
+                f.struct(
+                    f.col("nSamples").cast("integer").alias("sampleSize"),
+                    f.lit("European").alias("ancestry"),
                 )
-            )
-            .withColumn(
-                "ldPopulationStructure",
-                cls.aggregate_and_map_ancestries(f.col("discoverySamples")),
-            )
+            ),
+        ).withColumn(
+            "ldPopulationStructure",
+            cls.aggregate_and_map_ancestries(f.col("discoverySamples")),
         )
 
         return StudyIndex(