Merge branch 'main' into separate_spikeglx_catalogue_from_recording_probe

Author: h-mayorquin

.pre-commit-config.yaml

@@ -6,7 +6,7 @@ repos:
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
 -   repo: https://github.com/psf/black-pre-commit-mirror
-    rev: 25.11.0
+    rev: 25.12.0
     hooks:
     -   id: black
         files: ^src/|^tests/

resources/CN-logo.jpg

@@ -0,0 +1,188 @@
+"""
+2025-12-16 CambridgeNeurotech
+
+Derive probes to be used with SpikeInterface base on Cambridgeneurotech database at:
+https://github.com/cambridge-neurotech/probe_maps
+
+
+The output folder is ready to be used as a probeinterface library and contains:
+- one folder per probe
+- inside each folder a json file and a figure png file
+"""
+
+import argparse
+import json
+import shutil
+from pathlib import Path
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+from tqdm.auto import tqdm
+
+from probeinterface.plotting import plot_probe
+from probeinterface import write_probeinterface, Probe
+
+
+cn_logo = Path(__file__).parent / "CN-logo.jpg"
+
+parser = argparse.ArgumentParser(description="Generate CambridgeNeurotech probe library for probeinterface")
+parser.add_argument(
+    "probe_tables_path",
+    type=str,
+    help="Path to the folder containing the CambridgeNeurotech probe tables CSV files from https://github.com/cambridge-neurotech/probe_maps",
+)
+parser.add_argument(
+    "--output-folder", type=str, default="./cambridgeneurotech", help="Output folder to save the generated probes"
+)
+
+
+# graphing parameters
+plt.rcParams["pdf.fonttype"] = 42  # to make sure it is recognize as true font in illustrator
+plt.rcParams["svg.fonttype"] = "none"  # to make sure it is recognize as true font in illustrator
+
+
+def create_CN_figure(probe):
+    """
+    Create custom figire for CN with custom colors + logo
+    """
+    if probe.contact_sides is not None:
+        fig, axs = plt.subplots(ncols=2)
+        fig.set_size_inches(18.5, 10.5)
+    else:
+        fig, ax = plt.subplots()
+        fig.set_size_inches(18.5, 10.5)
+        axs = [ax]
+
+    n = probe.get_contact_count()
+    probe_max_height = np.max(probe.contact_positions[:, 1])
+    if probe.contact_sides is not None:
+        for i, side in enumerate(("front", "back")):
+            ax = axs[i]
+            plot_probe(
+                probe,
+                ax=ax,
+                contacts_colors=["#5bc5f2"] * n,  # made change to default color
+                probe_shape_kwargs=dict(
+                    facecolor="#6f6f6e", edgecolor="k", lw=0.5, alpha=0.3
+                ),  # made change to default color
+                with_contact_id=True,
+                side=side,
+            )
+            ax.set_title(f"Side: {side}", fontsize=20)
+    else:
+        plot_probe(
+            probe,
+            ax=axs[0],
+            contacts_colors=["#5bc5f2"] * n,  # made change to default color
+            probe_shape_kwargs=dict(
+                facecolor="#6f6f6e", edgecolor="k", lw=0.5, alpha=0.3
+            ),  # made change to default color
+            with_contact_id=True,
+        )
+        axs[0].set_title("")
+
+    for ax in axs:
+        y_min = ax.get_ylim()[0]
+        y_max = probe_max_height + 200
+        ax.set_ylim(y_min, y_max)
+        ax.set_xlabel("Width (\u03bcm)")  # modify to legend
+        ax.set_ylabel("Height (\u03bcm)")  # modify to legend
+        ax.spines["right"].set_visible(False)  # remove external axis
+        ax.spines["top"].set_visible(False)  # remove external axis
+
+    fig.suptitle("\n" + "CambridgeNeuroTech" + "\n" + probe.model_name, fontsize=24)
+
+    fig.tight_layout()
+
+    im = plt.imread(str(cn_logo))
+    newax = fig.add_axes([0.8, 0.85, 0.2, 0.1], anchor="NW", zorder=0)
+    newax.imshow(im)
+    newax.axis("off")
+
+    return fig
+
+
+def export_one_probe(probe_name, probe, output_folder):
+    """
+    Save one probe in "output_folder" + figure.
+    """
+    probe_folder = output_folder / probe_name
+    probe_folder.mkdir(exist_ok=True, parents=True)
+    probe_file = probe_folder / (probe_name + ".json")
+    figure_file = probe_folder / (probe_name + ".png")
+
+    write_probeinterface(probe_file, probe)
+
+    fig = create_CN_figure(probe)
+    fig.savefig(figure_file)
+
+    plt.close(fig)
+
+
+def is_contour_correct(probe):
+    from shapely.geometry import Point, Polygon
+
+    polygon = Polygon(probe.probe_planar_contour)
+
+    for i, contact_pos in enumerate(probe.contact_positions):
+        width = probe.contact_shape_params[i]["width"]
+        height = probe.contact_shape_params[i]["height"]
+        points = [
+            (contact_pos[0] - width / 2, contact_pos[1] - height / 2),
+            (contact_pos[0] + width / 2, contact_pos[1] - height / 2),
+            (contact_pos[0] + width / 2, contact_pos[1] + height / 2),
+            (contact_pos[0] - width / 2, contact_pos[1] + height / 2),
+        ]
+        for point in points:
+            p = Point(point[0], point[1])
+            if not polygon.contains(p):
+                return False
+    return True
+
+
+def generate_all_probes(probe_tables_path, output_folder):
+    sheet_names = list(pd.read_excel(probe_tables_path / "probe_contacts.xlsx", sheet_name=None).keys())
+
+    wrong_contours = []
+    sheets_with_issues = []
+
+    for sheet_name in tqdm(sheet_names, "Exporting CN probes"):
+        contacts = pd.read_excel(probe_tables_path / "probe_contacts.xlsx", sheet_name=sheet_name)
+        contour = pd.read_excel(probe_tables_path / "probe_contours.xlsx", sheet_name=sheet_name)
+
+        if np.all(pd.isna(contacts["contact_sides"])):
+            contacts.drop(columns="contact_sides", inplace=True)
+
+        if np.all(pd.isna(contacts["shank_ids"])):
+            contacts.drop(columns="shank_ids", inplace=True)
+
+        if "z" in contacts.columns:
+            contacts.drop(columns=["z"], inplace=True)
+        try:
+            probe = Probe.from_dataframe(contacts)
+            probe.manufacturer = "cambridgeneurotech"
+            probe.model_name = sheet_name
+            probe.set_planar_contour(contour)
+
+            if not is_contour_correct(probe):
+                wrong_contours.append(sheet_name)
+
+            export_one_probe(sheet_name, probe, output_folder)
+
+        except Exception as e:
+            print(f"Problem loading {sheet_name}: {e}")
+            sheets_with_issues.append(sheet_name)
+
+    print(f"\nREPORT for CambridgeNeurotech probe library generation\n")
+    print(f"Probes with wrong contours: {len(wrong_contours)}\n{wrong_contours}")
+    print(f"Probes failed to load: {len(sheets_with_issues)}\n{sheets_with_issues}")
+
+
+if __name__ == "__main__":
+    args = parser.parse_args()
+    probe_tables_path = Path(args.probe_tables_path)
+    output_folder = Path(args.output_folder)
+    if output_folder.exists():
+        shutil.rmtree(output_folder)
+    output_folder.mkdir(parents=True, exist_ok=True)
+    generate_all_probes(probe_tables_path, output_folder)