Skip to content

Commit 18a5b59

Browse files
dylansdanielsdylansdaniels
committed
text updates through end
1 parent 79ccbcc commit 18a5b59

File tree

4 files changed

+148
-290
lines changed

4 files changed

+148
-290
lines changed

‎coregui_gamma/coregui_gamma.html

Lines changed: 97 additions & 169 deletions
Original file line numberDiff line numberDiff line change
@@ -676,182 +676,110 @@ <h4 id="figure-20">Figure 20</h4>
676676
<p>In the next example, we will apply 50 Hz rhythmic synaptic inputs
677677
through proximal and distal projection patterns to produce gamma
678678
oscillations similar to those shown in Figure 8A of (Lee &amp; Jones,
679-
2013) [1]. In this simulation, the strength of the input is set so that
680-
the cells remain subthreshold and gamma rhythms emerge from subthreshold
679+
2013) [1].</p>
680+
<p>In this simulation, the strength of the input is set so that the
681+
cells remain subthreshold and gamma rhythms emerge from subthreshold
681682
current flow in the pyramidal neuron dendrites, rather than local
682683
spiking interactions as in the PING mechanisms described above.</p>
683-
<p>The parameter set that will simulate subthreshold gamma oscillations
684-
through rhythmic inputs to pyramidal neurons in proximal and distal
685-
projection patterns can be downloaded via the following hyperlink <a
686-
href="https://github.com/jonescompneurolab/hnn/blob/master/param/gamma_rhythmic_drive.param">gamma_rhythmic_drive.param</a>.
687-
This file can also be found in the HNN param subfolder.</p>
688-
<p>Load the parameter file values by returning to the main GUI window
689-
and clicking:</p>
690-
<pre><code>Set Parameters From File</code></pre>
691-
<p>Then select the file from HNN’s param subfolder or from your local
692-
machine. To view the parameters, click:</p>
693-
<pre><code>Set Parameters &gt; Rhythmic Proximal Inputs
694-
695-
Set Parameters &gt; Rhythmic Distal Inputs</code></pre>
696-
<p>You should see the values of adjustable parameters displayed in the
697-
dialog boxes below. In this example, the pyramidal neurons receive
698-
inputs, and the interneurons do not. The proximal and distal inputs
699-
start at 50.0 and 55.0 ms, respectively, and are slightly out of phase
700-
to allow synaptic inputs to effectively push current flow up the
701-
dendrites followed 5 ms later by current flow down the dendrites.
702-
Additionally, the input frequency for both proximal and distal inputs is
703-
set to a 50 Hz rhythm provided by “bursts” of excitatory synaptic input
704-
(driving burst frequencies; inter-burst-interval of 20 ms), with minimal
705-
noise within each driving burst (burst stdev of 2.5 ms). Please review
706-
the Alpha/Beta tutorial for a detailed description of the driving
707-
bursts. Note also that the amplitude of the inputs, which are only
708-
provided to the Layer 5 pyramidal neurons, is set to a small value (4e-5
709-
<span class="math inline"><em>μ</em><em>S</em></span>), which produces
710-
only subthreshold responses in the Layer 5 pyramidal neurons. As a
711-
result, the gamma mechanism shown in this simulation is fundamentally
712-
different from the previous examples, since it does not rely on the
713-
local-spiking interactions that underlie the PING mechanism.</p>
714-
<div class="stylefig" style="max-width: 800px;">
715-
<table>
716-
<h3>
717-
Figure 27
718-
</h3>
719-
<tr>
720-
<td>
721-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image2.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image2.png" alt="image2" />
722-
</a>
723-
</td>
724-
<td>
725-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image11.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image11.png" alt="image11" />
726-
</a>
727-
</td>
728-
<td>
729-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image47.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image47.png" alt="image47" />
730-
</a>
731-
</td>
732-
</tr>
733-
<tr>
734-
<td>
735-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image35.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image35.png" alt="image35" />
736-
</a>
737-
</td>
738-
<td>
739-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image5.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image5.png" alt="image5" />
740-
</a>
741-
</td>
742-
<td>
743-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image10.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image10.png" alt="image10" />
744-
</a>
745-
</td>
746-
</tr>
747-
</table>
748-
</div>
749-
<p>To run this simulation, return to the main GUI and click:</p>
750-
<pre><code>Start Simulation</code></pre>
751-
<p>This simulation runs for 550 ms of simulation time. Once completed,
752-
you will see output similar to that shown below.</p>
753-
<div class="stylefig" style="max-width: 850px;">
754-
<table>
755-
<h3>
756-
Figure 28
757-
</h3>
758-
<tr>
759-
<td>
760-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image40.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image40.png" alt="image40" width="100%"/>
761-
</a>
762-
</td>
763-
<td>
764-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image25.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image25.png" alt="image25" width="100%"/>
765-
</a>
766-
</td>
767-
</tr>
768-
</table>
769-
</div>
770-
<p>Histograms of the proximal and distal drives are shown in the top of
771-
the HNN GUI. The net dipole signal in this simulation shows a clear
772-
gamma rhythm at ~50 Hz, produced by the Layer 5 pyramidal neurons. Note,
773-
here the Layer 2/3 pyramidal neurons are not receiving any drive and
774-
hence do not contribute to the dipole current. There is only minor
775-
stochasticity to the synaptic inputs (burst stdev of 2.5 ms). Also note
776-
that the waveform shape in this simulation is distinct from the previous
777-
examples, lacking the sharp deflections produced by neuronal firing and
778-
strong somatic inhibition during PING.</p>
684+
<p>First, navigate to the <code>Network</code> tab and load the
685+
<code>gamma_rhythmic_drive</code> configuration from the
686+
<code>hnn_data</code> folder. Next, select the
687+
<code>External drives</code> tab and similarly load the subthreshold
688+
drives from the <code>gamma_rhythmic_drive</code> file. You will see two
689+
drives named <code>bursty1 (proximal)</code> and
690+
<code>bursty2 (distal)</code></p>
691+
<p>In this example, the pyramidal neurons receive inputs, and the
692+
interneurons do not. The proximal and distal inputs start at 50.0 and
693+
55.0 ms, respectively, and are slightly out of phase to allow synaptic
694+
inputs to effectively push current flow up the dendrites, followed 5 ms
695+
later by current flow down the dendrites. Additionally, the input
696+
frequency (as indicated by the <code>burst rate</code> parameter) for
697+
both proximal and distal inputs is set to a 50 Hz, representing “bursts”
698+
of excitatory synaptic input. The driving burst has an
699+
inter-burst-interval of 20 ms, with minimal noise within each driving
700+
burst (as indicated by the <code>Burst std dev</code> of 2.5 ms).</p>
701+
<p>:exclamation: For a detailed description of the driving bursts,
702+
please review the Alpha/Beta tutorial.</p>
703+
<p>Note also that the amplitude of the inputs, which are only provided
704+
to the Layer 5 pyramidal neurons, is set to a small value 0.00004, which
705+
produces only subthreshold responses in the Layer 5 pyramidal neurons.
706+
As a result, the gamma mechanism shown in this simulation is
707+
fundamentally different from the previous examples, since it does not
708+
rely on the local spiking interactions that underlie the PING
709+
mechanism.</p>
710+
<p>Next, change the <code>Name</code> in the <code>Simulation</code> tab
711+
to <code>gamma_rhythmic_drive</code> and click the <code>Run</code>
712+
button to run the simulation.</p>
713+
<p>Once completed, navigate to the <code>Visualization</code> tab and
714+
set the <code>Layout template</code> to <code>Dipole-Spectrogram</code>.
715+
Make sure you’ve sected the proper <code>Dataset</code>
716+
(<code>gamma_L5ping_L2ping</code>), and then click the
717+
<code>Make figure</code> button. Similarly, we’ll generate the PSD
718+
vizualization nby selecting <code>PSD Layers</code> from the
719+
<code>Layout template</code> and then clicking the
720+
<code>Make figure</code> button. You will see outputs similar to those
721+
shown in Figure 21 below.</p>
722+
<h4 id="figure-21.a">Figure 21.A</h4>
723+
<div style="display:block; width:90%; max-width:800px; margin: 0 auto;">
724+
<p><img src="images/gamma_fig_21_01.png" /></p>
725+
</div>
726+
<h4 id="figure-21.b">Figure 21.B</h4>
727+
<div style="display:block; width:90%; max-width:800px; margin: 0 auto;">
728+
<p><img src="images/gamma_fig_21_02.png" /></p>
729+
</div>
730+
<p>The net dipole signal in this simulation shows a clear gamma rhythm
731+
at ~50 Hz, produced by the Layer 5 pyramidal neurons. Note, here the
732+
Layer 2/3 pyramidal neurons are not receiving any drive and hence do not
733+
contribute to the dipole current. There is only minor stochasticity to
734+
the synaptic inputs (burst stdev of 2.5 ms). Also note that the waveform
735+
shape in this simulation is distinct from the previous examples, lacking
736+
the sharp deflections produced by neuronal firing and strong somatic
737+
inhibition during PING.</p>
779738
<h3
780-
id="gamma-through-rhythmicsubthreshold-synaptic-inputs-to-pyramidal-neurons-with-additional-noise">4.4
781-
Gamma through rhythmicsubthreshold synaptic inputs to pyramidal neurons,
782-
with additional noise</h3>
739+
id="gamma-through-rhythmic-subthreshold-synaptic-inputs-to-pyramidal-neurons-with-additional-noise">4.4
740+
Gamma through rhythmic subthreshold synaptic inputs to pyramidal
741+
neurons, with additional noise</h3>
783742
<p>In the final simulation of this section, we will add more noise to
784-
the previously applied 50 Hz rhythmic drive, described in section 4.2.
785-
The parameter set can be loaded by clicking the following hyperlink: <a
786-
href="https://github.com/jonescompneurolab/hnn/blob/master/param/gamma_rhythmic_drive_more_noise.param">gamma_rhythmic_drive_more_noise.param</a>.
787-
This file can also be found in the HNN’s param subfolder.</p>
788-
<p>To load the file, navigate to the main GUI window and click:</p>
789-
<pre><code>Set Parameters From File</code></pre>
790-
<p>Then select the file from HNN’s param subfolder or from your local
791-
machine. Next, view the parameters by clicking:</p>
792-
<pre><code>Set Parameters &gt; Rhythmic Proximal Inputs
793-
794-
Set Parameters &gt; Rhythmic Distal Inputs</code></pre>
795-
<p>You will see the values displayed in the dialog boxes below. Note
796-
that additional noise is present in the driving proximal and distal
797-
inputs, which now have Burst stdev = 5.0 ms, as compared to 2.5 ms in
798-
the previous simulations. This increase adds variability to timing of
799-
the burst of synaptic drive each input provides, and hence adds “noise”
800-
to the network.</p>
801-
<div class="stylefig" style="max-width: 650px;">
802-
<table>
803-
<h3>
804-
Figure 29
805-
</h3>
806-
<tr>
807-
<td>
808-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image13.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image13.png" alt="image13" />
809-
</a>
810-
</td>
811-
<td>
812-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image27.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image27.png" alt="image27" />
813-
</a>
814-
</td>
815-
</tr>
816-
</table>
817-
</div>
818-
<p>To run this simulation, navigate to the main GUI window and
819-
click:</p>
820-
<pre><code>Start Simulation</code></pre>
821-
<p>Once completed, you will see output similar to that shown below.</p>
822-
<div class="stylefig" style="max-width: 850px;">
823-
<table>
824-
<h3>
825-
Figure 30
826-
</h3>
827-
<tr>
828-
<td>
829-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image29.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image29.png" alt="image29" />
830-
</a>
831-
</td>
832-
<td>
833-
<a href="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image4.png"><img src="https://raw.githubusercontent.com/jonescompneurolab/hnn-tutorials/master/gamma/images/image4.png" alt="image4" />
834-
</a>
835-
</td>
836-
</tr>
837-
</table>
838-
</div>
839-
<p>In the simulation above, due to the higher variability in synaptic
840-
input timing, there is now more variability in the temporal dynamics and
841-
frequency content seen in the dipole signal. This is seen in the
842-
intermittent 50 Hz gamma events seen in the wavelet spectrogram. Here
843-
there is also a high power 25 Hz event that emerges, and both
844-
frequencies create peaks in the PSD from this single trial simulation,
845-
shown above.</p>
743+
the previously applied 50 Hz rhythmic drive.</p>
744+
<p>For each of the drives in the <code>External drives</code> tab,
745+
change the value of the <code>Burst std dev</code> from 2.5 ms to 5 ms.
746+
This increase adds variability to the timing of the burst of synaptic
747+
drive each input provides, and hence adds more “noise” to the
748+
network.</p>
749+
<p>Next, change the <code>Name</code> in the <code>Simulation</code> tab
750+
to <code>gamma_rhythmic_more_noise</code> and click the <code>Run</code>
751+
button to run the simulation.</p>
752+
<p>Once completed, navigate to the <code>Visualization</code> tab and
753+
set the <code>Layout template</code> to <code>Dipole-Spectrogram</code>.
754+
Make sure you’ve sected the proper <code>Dataset</code>, and then click
755+
the <code>Make figure</code> button. Similarly, we’ll generate the PSD
756+
vizualization by selecting <code>PSD Layers</code> from the
757+
<code>Layout template</code> and then clicking the
758+
<code>Make figure</code> button. You will see outputs similar to those
759+
shown in Figure 22 below.</p>
760+
<h4 id="figure-22.a">Figure 22.A</h4>
761+
<div style="display:block; width:90%; max-width:800px; margin: 0 auto;">
762+
<p><img src="images/gamma_fig_22_01.png" /></p>
763+
</div>
764+
<h4 id="figure-22.b">Figure 22.B</h4>
765+
<div style="display:block; width:90%; max-width:800px; margin: 0 auto;">
766+
<p><img src="images/gamma_fig_22_02.png" /></p>
767+
</div>
768+
<p>Due to the higher variability in synaptic input timing, there is now
769+
more variability in the temporal dynamics and frequency content seen in
770+
the dipole signal in the simulation above. This is seen in the
771+
intermittent 50 Hz gamma events visible in the wavelet spectrogram.</p>
772+
<p>With the added noise, we also observe lower-frequency peaks with
773+
relatively high power in the average PSD, though the intermittent gamma
774+
events still yield the highest-power peak observed in the PSD.</p>
846775
<h3 id="exercises-for-further-exploration-1">4.4.1 Exercises for further
847776
exploration</h3>
848777
<ul>
849-
<li><p>Go back to the gamma_L5weak_L2weak.param file and add recurrent
850-
synaptic connectivity between pyramidal neurons within a layer (e.g., L5
851-
Pyr -&gt; L5 Pyr weight = 9.1e-4(<span
852-
class="math inline"><em>μ</em><em>S</em></span>)); how does that
853-
influence the gamma rhythm? What happens as you change the strength of
854-
this connection?</p></li>
778+
<li><p>Go back to the <code>gamma_L5weak_L2weak</code> configuration and
779+
add recurrent synaptic connectivity between pyramidal neurons within a
780+
layer (e.g., L5 Pyr -&gt; L5 Pyr weight = 0.00091e-4); how does that
781+
change influence the gamma rhythm? What happens as you change the
782+
strength of this connection?</p></li>
855783
<li><p>Adjust the synaptic time constants of GABAA or other synapses;
856784
can you alter the peak gamma frequency?</p></li>
857785
<li><p>Add connectivity between Layer 2/3 and Layer 5? Is gamma

0 commit comments

Comments
 (0)