Merge pull request #6 from monarch-initiative/reports

Create report as a pivot table

Makefile

@@ -146,6 +146,11 @@ $(DATADIR)/curated-mondo-ncit-renal-subset.sssom.tsv: subsets/mondo-rare-renal.o
 #$(DATADIR)/curated-mondo-ncit-renal-subset.sssom.tsv:
 #	runoak -i sqlite:obo:mondo mappings .desc//p=i 'kidney disease' .and .desc//p=i 'hereditary disease' -O sssom -o $@
 
+###############################################
+############ Reports as Pivot table ###########
+###############################################
+$(RESULTDIR)/report_pivot_table.md:
+	python util/reports.py -o $@
 
 ###############################################
 ############ Ontologies #######################

results/report_pivot_table.md

@@ -0,0 +1,5 @@
+| subject_prefix   |   MmusDv |   NCIT |   WBbt |   ZFA |
+|:-----------------|---------:|-------:|-------:|------:|
+| FBbt             |        0 |      0 |     41 |    72 |
+| HsapDv           |       22 |      0 |      0 |     0 |
+| MONDO            |        0 |     25 |      0 |     0 |

util/reports.py

@@ -0,0 +1,56 @@
+import click
+from pathlib import Path
+import glob
+import pandas as pd
+PROJECT_DIR = Path(__file__).resolve().parents[1]
+DATA_DIR = PROJECT_DIR / "data"
+RESULTS_DIR = PROJECT_DIR / "results"
+SUBJECT_ID = "subject_id"
+OBJECT_ID = "object_id"
+PREDICATE_ID = "predicate_id"
+
+
+@click.group()
+def main():
+    pass
+
+@main.command()
+@click.option("-o", "--outfile", default=RESULTS_DIR / "report_pivot_table.md")
+def reports(outfile: str):
+    """Run reports."""
+    # import all tsv files using pandas and just 3 columns
+    tsv_files = glob.glob(str(DATA_DIR / "*.tsv"))
+
+    # Read and concatenate dataframes in one step
+    dfs = (pd.read_csv(file, sep='\t', usecols=['subject_id', 'predicate_id', 'object_id'], comment="#") for file in tsv_files)
+    large_df = pd.concat(dfs, ignore_index=True).drop_duplicates()
+
+    # Split subject_id and object_id
+    large_df[['subject_prefix', 'subject_id']] = large_df['subject_id'].str.split(":", expand=True)
+    large_df[['object_prefix', 'object_id']] = large_df['object_id'].str.split(":", expand=True)
+
+    # unique_object_prefixes = list(large_df['object_prefix'].unique())
+    # unique_subject_prefixes = list(large_df['subject_prefix'].unique())
+
+    # # Group by subject_prefix and object_prefix
+    # grouped_df = large_df.groupby(['subject_prefix', 'object_prefix'])
+
+    # mondo_ncit_df = grouped_df.get_group(('MONDO', 'NCIT')).drop_duplicates()
+    # hsapdv_mmusdv_df = grouped_df.get_group(('HsapDv', 'MmusDv')).drop_duplicates()
+    # fbbt_wbbt_df = grouped_df.get_group(('FBbt', 'WBbt')).drop_duplicates()
+    # fbbt_zfa_df = grouped_df.get_group(('FBbt', 'ZFA')).drop_duplicates()
+
+    # Create a pivot table
+    pivot_table = large_df.pivot_table(index='subject_prefix', columns='object_prefix', aggfunc='size', fill_value=0)
+
+    # Convert the pivot table to markdown
+    markdown_table = pivot_table.to_markdown()
+
+    # Write the markdown table to a file
+    with open(str(outfile), 'w') as f:
+        f.write(markdown_table)
+
+
+
+if __name__ == "__main__":
+    reports()