Merge pull request #162 from singjc/update/export_scored_report

Update/export scored report

Author: singjc

pyprophet/export/export_report.py

@@ -1,7 +1,10 @@
 import sqlite3
 import pandas as pd
 
+
+from .._config import ExportIOConfig
 from ..report import post_scoring_report
+from ..io.dispatcher import ReaderDispatcher
 from ..io.util import get_parquet_column_names
 from ..io.util import check_sqlite_table
 from ..report import plot_scores
@@ -130,35 +133,19 @@ def export_scored_report(
         The format of the scoring report, either 'osw' or 'parquet'. Default is 'osw'.
     """
 
-    cols_infile = get_parquet_column_names(infile)
-
-    select_cols = [
-        "RUN_ID",
-        "PROTEIN_ID",
-        "PEPTIDE_ID",
-        "PRECURSOR_ID",
-        "PRECURSOR_DECOY",
-        "FEATURE_MS2_AREA_INTENSITY",
-        "SCORE_MS2_SCORE",
-        "SCORE_MS2_PEAK_GROUP_RANK",
-        "SCORE_MS2_Q_VALUE",
-        "SCORE_PEPTIDE_GLOBAL_SCORE",
-        "SCORE_PEPTIDE_GLOBAL_Q_VALUE",
-        "SCORE_PEPTIDE_EXPERIMENT_WIDE_SCORE",
-        "SCORE_PEPTIDE_EXPERIMENT_WIDE_Q_VALUE",
-        "SCORE_PEPTIDE_RUN_SPECIFIC_SCORE",
-        "SCORE_PEPTIDE_RUN_SPECIFIC_Q_VALUE",
-        "SCORE_PROTEIN_GLOBAL_SCORE",
-        "SCORE_PROTEIN_GLOBAL_Q_VALUE",
-        "SCORE_PROTEIN_EXPERIMENT_WIDE_SCORE",
-        "SCORE_PROTEIN_EXPERIMENT_WIDE_Q_VALUE",
-        "SCORE_IPF_QVALUE",
-    ]
+    config = ExportIOConfig(
+        infile=infile,
+        outfile=outfile,
+        subsample_ratio=1.0,  # not used for export/report paths
+        level="export",
+        context="export_scored_report",
+        # no need to set export_format for this utility
+    )
 
-    # Filter select cols based on available columns in the input file
-    select_cols = [col for col in select_cols if col in cols_infile]
+    # Get the right reader for the detected file type & context.
+    reader = ReaderDispatcher.get_reader(config)
 
-    # Load the input data
-    df = pd.read_parquet(infile, columns=select_cols)
+    # Read once (works for OSW or Parquet via their respective readers).
+    df = reader.read()
 
     post_scoring_report(df, outfile)