feat(morph-plot): also allow plotting of swc files

Note that we first export the SWC files to NeuroML and then visualise
them, we don't visualise them "natively". That will require the
implementation of new methods to create meshes from SWC points,
effectively re-doing the SWC to NeuroML conversion logic.

Author: sanjayankur31

pyneuroml/plot/PlotMorphology.py

@@ -12,6 +12,7 @@
 import os
 import random
 import sys
+import textwrap
 import typing
 from typing import Optional
 
@@ -23,6 +24,7 @@
 from neuroml.utils import fix_external_morphs_biophys_in_cell
 
 from pyneuroml.pynml import read_neuroml2_file
+from pyneuroml.swc.ExportNML import convert_swc_to_neuroml
 from pyneuroml.utils import extract_position_info
 from pyneuroml.utils.cli import build_namespace
 from pyneuroml.utils.plot import (
@@ -46,14 +48,25 @@ def process_args():
     Parse command-line arguments.
     """
     parser = argparse.ArgumentParser(
-        description=("A script which can generate plots of morphologies in NeuroML 2")
+        description=(
+            textwrap.dedent(
+                """
+            A script which can generate plots of morphologies in NeuroML 2.
+            It can generate plots of single cells (either detailed or point
+            cells), and networks that include these cells.
+
+            It can also visualise single cell morphologies from SWC files,
+            although it will first convert these to NeuroML.
+            """
+            )
+        )
     )
 
     parser.add_argument(
         "nmlFile",
         type=str,
-        metavar="<NeuroML 2 file>",
-        help="Name of the NeuroML 2 file",
+        metavar="<NeuroML 2 file/SWC cell file>",
+        help="Name of the NeuroML 2 file/SWC cell file",
     )
 
     parser.add_argument(
@@ -215,17 +228,20 @@ def plot_2D(
     If a file with a network containing multiple cells is provided, it will
     plot all the cells. For detailed neuroml.Cell types, it will plot their
     complete morphology. For point neurons, we only plot the points (locations)
-    where they are.
+    where they are. For single cell SWC files, it will first convert them to
+    NeuroML and then plot them.
 
     This method uses matplotlib.
 
     .. versionadded:: 1.1.12
         The hightlight_spec parameter
 
 
-    :param nml_file: path to NeuroML cell file, or a NeuroMLDocument object
-    :type nml_file: str or :py:class:`neuroml.NeuroMLDocument` or
-        :py:class:`neuroml.Cell`
+    :param nml_file: path to NeuroML cell file or single cell SWC file or
+        :py:class:`neuroml.NeuroMLDocument` or :py:class:`neuroml.Cell`
+        or :py:class:`neuroml.Morphology` object
+    :type nml_file: str or neuroml.NeuroMLDocument or neuroml.Cell or
+        neuroml.Morphology
     :param plane2d: what plane to plot (xy/yx/yz/zy/zx/xz)
     :type plane2d: str
     :param min_width: minimum width for segments (useful for visualising very
@@ -317,6 +333,14 @@ def plot_2D(
         # TODO: check if this is required: must for MultiscaleISN
         if nml_file.endswith(".h5"):
             nml_model = read_neuroml2_file(nml_file)
+        elif nml_file.endswith(".swc"):
+            nml_model_doc = convert_swc_to_neuroml(
+                nml_file,
+                neuroml_file=None,
+                standalone_morphology=False,
+                unbranched_segment_groups=False,
+            )
+            nml_model = nml_model_doc.cells[0]
         else:
             nml_model = read_neuroml2_file(
                 nml_file,

pyneuroml/plot/PlotMorphologyVispy.py

@@ -34,6 +34,7 @@
 from scipy.spatial.transform import Rotation
 
 from pyneuroml.pynml import read_neuroml2_file
+from pyneuroml.swc.ExportNML import convert_swc_to_neuroml
 from pyneuroml.utils import extract_position_info, make_cell_upright
 from pyneuroml.utils.plot import (
     DEFAULTS,
@@ -361,8 +362,13 @@ def plot_interactive_3D(
     .. versionadded:: 1.1.12
         The highlight_spec parameter
 
+    If a file with a network containing multiple cells is provided, it will
+    plot all the cells. For detailed neuroml.Cell types, it will plot their
+    complete morphology. For point neurons, we only plot the points (locations)
+    where they are as spheres. For single cell SWC files, it will first convert
+    them to NeuroML and then plot them.
 
-    :param nml_file: path to NeuroML cell file or
+    :param nml_file: path to NeuroML cell file or single cell SWC file or
         :py:class:`neuroml.NeuroMLDocument` or :py:class:`neuroml.Cell`
         or :py:class:`neuroml.Morphology` object
     :type nml_file: str or neuroml.NeuroMLDocument or neuroml.Cell or
@@ -502,6 +508,14 @@ def plot_interactive_3D(
         # TODO: check if this is required: must for MultiscaleISN
         if nml_file.endswith(".h5"):
             nml_model = read_neuroml2_file(nml_file)
+        elif nml_file.endswith(".swc"):
+            nml_model_doc = convert_swc_to_neuroml(
+                nml_file,
+                neuroml_file=None,
+                standalone_morphology=False,
+                unbranched_segment_groups=False,
+            )
+            nml_model = nml_model_doc.cells[0]
         else:
             # do not fix external morphs here, we do it later below
             nml_model = read_neuroml2_file(
@@ -742,6 +756,7 @@ def plot_interactive_3D(
             except KeyError:
                 # if single cell only, use default groups
                 if total_cells == 1:
+                    logger.debug("Only one detailed cell, using default color groups")
                     color = "default groups"
                 # if multiple cells, use different colors for each cell
                 else:
@@ -917,15 +932,12 @@ def plot_3D_cell_morphology(
 
     .. seealso::
 
-        :py:func:`plot_2D`
+        :py:func:`plot_interactive_3D`
             general function for plotting
 
-        :py:func:`plot_2D_schematic`
+        :py:func:`plot_3D_schematic`
             for plotting only segmeng groups with their labels
 
-        :py:func:`plot_2D_point_cells`
-            for plotting point cells
-
     :param offset: offset for cell
         Note that this is only used in creating the meshdata. If set to None,
         this is ommitted from the meshdata, and the mesh creator will set it to