[MRG] Record calcium concentration #792

katduecker · 2024-06-12T20:52:20Z

Recording calcium concentration in all sections of the L5 pyramidal neurons

ntolley · 2024-06-12T22:48:51Z

hnn_core/cell.py

+                if 'ca' in self.sections[sec_name].mechs:
+                    self.ca[sec_name] = h.Vector()
+                    self.ca[sec_name].record(
+                        self._nrn_sections[sec_name](0.5)._ref_cai) # there may be an issue here with referencing mod file


keep or remove?

jasmainak · 2024-06-13T02:54:36Z

hnn_core/cell.py

@@ -348,6 +348,11 @@ class Cell:
        by synapse type (keys can be soma_gabaa, soma_gabab etc.).
        Must be enabled by running simulate_dipole(net, record_isec=True)
        or simulate_dipole(net, record_isoma=True)
+    ca : dict
+        Contains recording of section speicifc calcium concentration. 
+        Must be enabled by running simulate_dipole(...) with record_ca


Suggested change

Must be enabled by running simulate_dipole(...) with record_ca

Must be enabled by running simulate_dipole(net, record_ca)

Shouldn't we tell the user what to do? Because "True" won't work

I think @jasmainak is copying that from the docstring above for record_isoma, which is out of date.

But yes the docstring should read Must be enabled by running simulate_dipole(..., record_ca='all')

I just wanted to say that use the function signature rather than using words to explain it.

jasmainak · 2024-06-13T03:03:34Z

hnn_core/tests/test_dipole.py

    assert len(mpi_net.cell_response.vsec) == n_trials
    assert len(mpi_net.cell_response.isec) == n_trials
    assert len(mpi_net.cell_response.vsec[trial_idx][gid]) == 8  # num sec
-    assert len(mpi_net.cell_response.isec[trial_idx][gid]) == 1
+    assert len(mpi_net.cell_response.isec[trial_idx][gid]) == 1  


this must be a pep8 issue

jasmainak · 2024-06-13T03:04:17Z

@ntolley monthly issue metrics fails !

ntolley · 2024-06-13T12:42:16Z

@ntolley monthly issue metrics fails !

Fixed!

ntolley · 2024-06-13T13:13:20Z

hnn_core/dipole.py

@@ -37,6 +37,9 @@ def simulate_dipole(net, tstop, dt=0.025, n_trials=None, record_vsec=False,
    record_isec : 'all' | 'soma' | False
        Option to record voltages from all sections ('all'), or just
        the soma ('soma'). Default: False.
+    record_ca : 'all' | 'soma' | False


You can actually just copy the docstring here to the one in cell.py

Can you also replace the incorrect record isoma/vsoma docstrings in cell.py with the isec/vsec docstrings here?

ntolley · 2024-06-13T13:18:32Z

hnn_core/cell.py

+
+        if record_ca:     
+            for sec_name in self.ca:
+                if 'ca' in self.sections[sec_name].mechs:


Suggested change

if 'ca' in self.sections[sec_name].mechs:

if hasattr(self._nrn_sections[sec_name](0.5), '_ref_cai)::

Your suggestion! Just making sure we don't forget

jasmainak

PR looks clean to me. I leave it to @ntolley to check and merge :) Don't forget to update whats_new !

ntolley · 2024-06-14T14:09:21Z

Here's the plot with the spikes/Ca overlaid. At first I thought something was wrong but realized I hadn't set the times correctly.

Very nice work @katduecker! This definitely looks good

from hnn_core import simulate_dipole, jones_2009_model
import numpy as np
import matplotlib.pyplot as plt

net = jones_2009_model()
net.add_evoked_drive('evprox', mu=20.0, sigma=2.0, numspikes=1,
                     location='proximal', weights_ampa={'L5_pyramidal': 0.1})

plt.figure()
fig, ax = plt.subplots(2, 1, figsize=(5, 4), sharex=True, tight_layout=True)
plot_data = np.array([net.cell_response.ca[0][gid]['soma'] for gid in net.gid_ranges['L5_pyramidal']]).T
ax[0].plot(dpl[0].times, plot_data)
ax[0].set_title('Somatic Ca$^{2+}$')
ax[1].set_title('Spike Raster')
net.cell_response.plot_spikes_raster(ax=ax[1])

katduecker · 2024-06-19T01:24:19Z

Thanks both! I haven't managed to look at the rest of this yet, as I have been busy with the NEURON workshop. One thing I learned today at the workshop is that calcium concentration should be at the order a few hundred nM, but our calcium concentration is at the order of 0.1 mM (mod file of ca shows that cai is in mM). This is 1000 times too high... Am I missing something obvious with a conversion of the units?

jasmainak · 2024-06-19T01:59:49Z

We use reduced neurons acc to Bush and Sejnowski 93, so do we actually expect physical units?

…

Sent from my iPhone

On Tue 18 Jun 2024 at 21:24, Katharina Duecker ***@***.***> wrote: Thanks both! I haven't managed to look at the rest of this yet, as I have been busy with the NEURON workshop. One thing I learned today at the workshop is that calcium concentration should be at the order a few hundred nM, but our calcium concentration is at the order of 0.1 mM (mod file of ca shows that cai is in mM). This is 1000 times too high... Am I missing something obvious with a conversion of the units? — Reply to this email directly, view it on GitHub <#792 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADY6FIUFFDSYUXZ36CR7SHLZIDMVTAVCNFSM6AAAAABJHDF536VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNZXGM2DOOJRGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

katduecker · 2024-06-19T14:20:51Z

I asked Stephanie about this and she said that it may be possible that the conductance had to be scaled up to get the desired dynamics. To confirm this, I will read out the concentration for sodium and potassium as well, to see if they are in a similar regime. I remember @ntolley saying that an option to read out concentrations for all ions may not be a useful functionality in general, but if it allows sanity checks of the scale of the conductance, couldn't it be helpful for users to have access to this? I'd be happy to write a tutorial on this if we decide to integrate.

katduecker added 2 commits June 12, 2024 16:26

record calcium concentration in L5pyr

d3105f9

fixed unit tests calcium L5

aa9f591

ntolley reviewed Jun 12, 2024

View reviewed changes

jasmainak reviewed Jun 13, 2024

View reviewed changes

fix pep8

33af4e4

ntolley reviewed Jun 13, 2024

View reviewed changes

jasmainak approved these changes Jun 13, 2024

View reviewed changes

ntolley changed the title ~~Record calcium concentration~~ [MRG] Record calcium concentration Jun 14, 2024

katduecker closed this Jun 25, 2024

katduecker deleted the record_calc branch June 25, 2024 17:02

katduecker mentioned this pull request Jun 25, 2024

Record calcium concentration new #804

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[MRG] Record calcium concentration #792

[MRG] Record calcium concentration #792

katduecker commented Jun 12, 2024

ntolley Jun 12, 2024

jasmainak Jun 13, 2024 •

edited

Loading

katduecker Jun 13, 2024 •

edited

Loading

ntolley Jun 13, 2024

jasmainak Jun 13, 2024

jasmainak Jun 13, 2024

jasmainak commented Jun 13, 2024

ntolley commented Jun 13, 2024

ntolley Jun 13, 2024

ntolley Jun 13, 2024

ntolley Jun 13, 2024

jasmainak left a comment

ntolley commented Jun 14, 2024 •

edited

Loading

katduecker commented Jun 19, 2024

jasmainak commented Jun 19, 2024 via email

katduecker commented Jun 19, 2024 •

edited

Loading

	Must be enabled by running simulate_dipole(...) with record_ca
	Must be enabled by running simulate_dipole(net, record_ca)

	if 'ca' in self.sections[sec_name].mechs:
	if hasattr(self._nrn_sections[sec_name](0.5), '_ref_cai)::

[MRG] Record calcium concentration #792

[MRG] Record calcium concentration #792

Conversation

katduecker commented Jun 12, 2024

ntolley Jun 12, 2024

Choose a reason for hiding this comment

jasmainak Jun 13, 2024 • edited Loading

Choose a reason for hiding this comment

katduecker Jun 13, 2024 • edited Loading

Choose a reason for hiding this comment

ntolley Jun 13, 2024

Choose a reason for hiding this comment

jasmainak Jun 13, 2024

Choose a reason for hiding this comment

jasmainak Jun 13, 2024

Choose a reason for hiding this comment

jasmainak commented Jun 13, 2024

ntolley commented Jun 13, 2024

ntolley Jun 13, 2024

Choose a reason for hiding this comment

ntolley Jun 13, 2024

Choose a reason for hiding this comment

ntolley Jun 13, 2024

Choose a reason for hiding this comment

jasmainak left a comment

Choose a reason for hiding this comment

ntolley commented Jun 14, 2024 • edited Loading

katduecker commented Jun 19, 2024

jasmainak commented Jun 19, 2024 via email

katduecker commented Jun 19, 2024 • edited Loading

jasmainak Jun 13, 2024 •

edited

Loading

katduecker Jun 13, 2024 •

edited

Loading

ntolley commented Jun 14, 2024 •

edited

Loading

katduecker commented Jun 19, 2024 •

edited

Loading