crispr data/fig: filter casc results (issue #19)

figures/figures.smk

@@ -164,7 +164,7 @@ rule fig_crispr_data:
         import os
         import re
         import pandas
-        from src.utils import parse_minced_report
+        from src.utils import parse_minced_report, parse_casc_report
         summary = []
         for ifile_path in input:
             ifile_df = None
@@ -177,11 +177,9 @@ rule fig_crispr_data:
             asm_tool = os.path.basename(ifile_path).split(".")[0]
             # file content
             if crispr_tool == "minced":
-                ifile_df = parse_minced_report(ifile_path)
-                ifile_df = pandas.DataFrame(ifile_df)
+                ifile_df = pandas.DataFrame(parse_minced_report(ifile_path))
             elif crispr_tool == "casc":
-                ifile_df = pandas.read_csv(ifile_path, sep="\t", header=0, usecols=["#seq_id", "spacers", "array_start", "array_stop"], comment=None)
-                ifile_df.rename(columns={"#seq_id": "seq_id"}, inplace=True)
+                ifile_df = parse_casc_report(ifile_path)
             ifile_df = ifile_df.assign(crispr_tool=crispr_tool)
             ifile_df = ifile_df.assign(asm_tool=asm_tool)
             summary.append(ifile_df)

figures/src/utils.py

 #!/usr/bin/python
 
+def parse_casc_report(ifile_path):
+    import pandas
+    # read in
+    summary = pandas.read_csv(ifile_path, sep="\t", header=0, usecols=["#seq_id", "spacers", "array_start", "array_stop", "bonafide"], comment=None)
+    # rename columns
+    summary.rename(columns={"#seq_id": "seq_id"}, inplace=True)
+    # filter
+    summary = summary[summary["bonafide"]]
+    assert all(summary["bonafide"]) # just a check
+    # drop columns
+    summary.drop(columns=["bonafide"], inplace=True)
+    return summary
+
 def parse_minced_report(ifile_path):
     """
     Parse the MINCED report: for each sequence and found CRISPR array, extract