Using Single Neuron Dynamics to Predict Synchronous Global Network - - PowerPoint PPT Presentation

▶
using single neuron dynamics to predict synchronous
SMART_READER_LITE
LIVE PREVIEW

Using Single Neuron Dynamics to Predict Synchronous Global Network - - PowerPoint PPT Presentation

Mar 20, 2024 461 likes •766 views

Using Single Neuron Dynamics to Predict Synchronous Global Network Activities Robert Kim Neurosciences Graduate Program University of California, San Diego December 8, 2017 Neurodynamics Final Project December 8, 2017 1 / 16 Outline

slide-1
SLIDE 1

Using Single Neuron Dynamics to Predict Synchronous Global Network Activities

Robert Kim

Neurosciences Graduate Program University of California, San Diego

December 8, 2017

Neurodynamics Final Project December 8, 2017 1 / 16

slide-2
SLIDE 2

Outline

1

Background

2

Methods

3

Results

4

Summary

Neurodynamics Final Project December 8, 2017 2 / 16

slide-3
SLIDE 3

Outline

1

Background

2

Methods

3

Results

4

Summary

Neurodynamics Final Project December 8, 2017 3 / 16

slide-4
SLIDE 4

Background

Based on the previous findings from

Neurodynamics Final Project December 8, 2017 4 / 16

slide-5
SLIDE 5

Background

Based on the previous findings from Rat cortical neurons artificially grown on a multi-electrode arrays

(Figure from Tajima, S. et al. Locally embedded presages of global network bursts. PNAS, 114: 9517-9522, 2017) Neurodynamics Final Project December 8, 2017 4 / 16

slide-6
SLIDE 6

Background

10 15 20 25 30

Time (sec)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Mean Firing Activity

5

Neurodynamics Final Project December 8, 2017 5 / 16

slide-7
SLIDE 7

Background

10 15 20 25 30

Time (sec)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Mean Firing Activity

5

1 0.2 0.4 0.8 0.6 0.8 0.6 0.9 1 0.8 0.4 0.7

x(t)

0.6 0.5 0.4 0.2 0.3 0.2 0.1

x1(t) x2(t) (Figure from Tajima, S. et al. Locally embedded presages of global network bursts. PNAS, 114: 9517-9522, 2017) Neurodynamics Final Project December 8, 2017 5 / 16

slide-8
SLIDE 8

Outline

1

Background

2

Methods

3

Results

4

Summary

Neurodynamics Final Project December 8, 2017 6 / 16

slide-9
SLIDE 9

Methods

Izhikevich neurons will be used to replicate some of the findings of Tajima et al. ˙ v = 0.04v2 + 5v + 140 − u + I ˙ u = a · (bv − u) if v ≥ 35 mV, v = c and u = u + d

Neurodynamics Final Project December 8, 2017 7 / 16

slide-10
SLIDE 10

Methods

Izhikevich neurons will be used to replicate some of the findings of Tajima et al. ˙ v = 0.04v2 + 5v + 140 − u + I ˙ u = a · (bv − u) if v ≥ 35 mV, v = c and u = u + d

  • 100
  • 50

50 100 150 200 250 300 350 400

Time (ms)

  • 80
  • 60
  • 40
  • 20

20

v (mV)

  • 100
  • 50

50 100 150 200 250 300 350 400

Time (ms)

  • 80
  • 60
  • 40
  • 20

20

v (mV)

  • 100
  • 50

50 100 150 200 250 300 350 400

Time (ms)

  • 80
  • 60
  • 40
  • 20

20

v (mV)

  • 100
  • 50

50 100 150 200 250 300 350 400

Time (ms)

  • 100
  • 50

v (mV)

Neurodynamics Final Project December 8, 2017 7 / 16

slide-11
SLIDE 11

Methods

RS Excitatory Neuron a = 0.02, b = 0.13 c = −65 mV, d = 8

200 400 600 800 1000

Time (ms)

  • 80
  • 60
  • 40
  • 20

20

Voltage (mV)

Neurodynamics Final Project December 8, 2017 8 / 16

slide-12
SLIDE 12

Methods

RS Excitatory Neuron a = 0.02, b = 0.13 c = −65 mV, d = 8

200 400 600 800 1000

Time (ms)

  • 80
  • 60
  • 40
  • 20

20

Voltage (mV)

FS Inhibitory Neuron a = 0.1, b = 0.13 c = −65 mV, d = 2

200 400 600 800 1000

Time (ms)

  • 60
  • 40
  • 20

20

Voltage (mV)

Neurodynamics Final Project December 8, 2017 8 / 16

slide-13
SLIDE 13

Methods

Network Configuration 450 Izhikevich neurons

360 RS excitatory neurons 90 FS inhibitory neurons

Neurodynamics Final Project December 8, 2017 9 / 16

slide-14
SLIDE 14

Methods

Network Configuration 450 Izhikevich neurons

360 RS excitatory neurons 90 FS inhibitory neurons

Sparsely and randomly connected to one another (40% connections)

Neurodynamics Final Project December 8, 2017 9 / 16

slide-15
SLIDE 15

Methods

Network Configuration 450 Izhikevich neurons

360 RS excitatory neurons 90 FS inhibitory neurons

Sparsely and randomly connected to one another (40% connections) No external/inject current

Neurodynamics Final Project December 8, 2017 9 / 16

slide-16
SLIDE 16

Outline

1

Background

2

Methods

3

Results

4

Summary

Neurodynamics Final Project December 8, 2017 10 / 16

slide-17
SLIDE 17

Results

130 131 132 133 134 135 136

Time (sec)

130 131 132 133 134 135 136

Time (sec)

1

Mean Firing Activity Neurons

Neurodynamics Final Project December 8, 2017 11 / 16

slide-18
SLIDE 18

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-19
SLIDE 19

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-20
SLIDE 20

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-21
SLIDE 21

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-22
SLIDE 22

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-23
SLIDE 23

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-24
SLIDE 24

Results

Dynamics of individual neurons were able to forecast/predict the spontaneous global burst.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

0.2 0.4 0.6 0.8 1

x(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x1(t) Neurodynamics Final Project December 8, 2017 12 / 16

slide-25
SLIDE 25

Results

Single Neuron Dynamics Forecast

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t)

Mean Population Dynamics Forecast

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

b(t)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

b1(t) Neurodynamics Final Project December 8, 2017 13 / 16

slide-26
SLIDE 26

Outline

1

Background

2

Methods

3

Results

4

Summary

Neurodynamics Final Project December 8, 2017 14 / 16

slide-27
SLIDE 27

Summary

Main findings of Tajima et al. were also observed in a network of Izhikevich neurons Future work:

Investigate the synaptic connectivity and strength of “good predictor” neurons Investigate the effects of sparsity and E/I balance on predictability Investigate the effects of synaptic strength on predictability

Neurodynamics Final Project December 8, 2017 15 / 16

slide-28
SLIDE 28

Acknowledgement

Gerald Pao, MD, PhD

Neurodynamics Final Project December 8, 2017 16 / 16

slide-29
SLIDE 29

Acknowledgement

Gerald Pao, MD, PhD

Any questions?

Neurodynamics Final Project December 8, 2017 16 / 16