tool to find pairs of cells: path to a csv file is harcoded

ephys/tools/find_cell_pairs.py

@@ -1,17 +1,68 @@
 import pandas as pd
 from pathlib import Path
+import datetime
+import ephys.tools.filename_tools as FT
 
-    """ Find cell pairs with the following criteria:
-    1. 2 different cell types, same slice
-    2. 2 different cell types, different slices, same day
-    3. 2 different cell types, different slices, different days, but with similar ages
+""" Find cell pairs with the following criteria:
+1. 2 different cell types, same slice
+2. 2 different cell types, different slices, same day
+3. 2 different cell types, different slices, different days, but with similar ages
+4. same cell type in the same slice.
+"""
 
-    """
 
-files  = Path("R_statisc_sumaries/rmtau_05-feb02025.csv")
-print("File exists: ", files.exists())
-df = pd.read_csv(files)
+def find_pairs(celltype1_name, celltype2_name, csv_file:str, diff_slice:bool=False, diff_date:bool=False):
+    files = Path(csv_file)  # Path("R_statistics_summaries/rmtau_05-Feb-2025.csv")
+    print("File exists: ", files.exists())
+    df = pd.read_csv(files)
+    print(df.columns)
 
-giants = df[df["cell_type"] == "giant"]
-pyramids = df[df["cell_type"] == "pyramidal"]
-print("Giants: ", giants)
+    celltype1 = df[df["cell_type"] == celltype1_name]
+    celltype2 = df[df["cell_type"] == celltype2_name]
+
+    for row in celltype1.itertuples():
+        cell_id = row.cell_id
+        date = cell_id.split("_")[0]
+        celltype2_on_date = celltype2[celltype2["cell_id"].str.startswith(date)]
+        celltype1_id = FT.make_cellid_from_slicecell(cell_id)
+        celltype1_slice = Path(celltype1_id).parts[1]
+        if len(celltype2_on_date) :
+
+            for celltype2_cell in celltype2_on_date.itertuples():
+                celltype2_id = FT.make_cellid_from_slicecell(celltype2_cell.cell_id)
+                sliceno = Path(celltype2_id).parts[1]
+                if sliceno == celltype1_slice and celltype2_id != celltype1_id:
+                    print(
+                    f"Celltype 1 {celltype1_name:s}: {celltype1_id}, P{int(row.age):d} has these pyramidal cells to match: "
+                        )
+                    print(f"          Same Slice: {celltype2_id:s}, P{int(celltype2_cell.age):d}\n")
+                else:
+                    if diff_slice:
+                        print(f"         Diff Slice: {celltype2_id:s}, P{int(celltype2_cell.age):d}")
+        else:
+            if not diff_date:
+                continue
+            # Search for pyr cells with same age, sorted by nearby dates
+            celltype1_datetime = datetime.datetime.strptime(date, "%Y.%m.%d")
+            print(
+                f"{celltype1_name:s} cell: {celltype1_id} P{int(row.age):d} has these pyramidal cells to match: "
+            )
+            for celltype2 in celltype2_on_date.itertuples():
+                celltype2_date = datetime.datetime.strptime(
+                    celltype2_on_date.cell_id.split("_")[0], "%Y.%m.%d"
+                )
+                daydiff = abs((celltype1_datetime - celltype2_date).days)
+                if daydiff < 10 and daydiff > 0 and abs(row.age - celltype2_on_date.age) < 5:
+                    print(
+                        f"    Close date: {celltype2_on_date.cell_id} {daydiff:d} days apart and age difference: {int(abs(row.age - celltype2_on_date.age)):d}"
+                    )
+
+            # print(f"Giant cell: {giant_id} P{int(row.age):d} has no pyramidal cells to match")
+            # print("Found a pair, same date: ", cell_id, pyrs_on_date.iloc[0].cell_id)
+
+if __name__ == "__main__":
+    celltype1_name = "pyramidal"
+    celltype2_name = "pyramidal"
+    csv_file = "/Users/pbmanis/Desktop/Python/mrk-nf107/rmtau_12-Feb-2025.csv"
+
+    find_pairs(celltype1_name, celltype2_name, csv_file, diff_slice=False, diff_date=False)