Does the usage of the Shape may affect mV readouts?

When Python is the interpreter, what is a good
design for the interface to the basic NEURON
concepts.

Moderator: hines

Post Reply
ziemek
Posts: 43
Joined: Thu May 23, 2019 8:02 am
Location: Warsaw, Poland
Contact:

Does the usage of the Shape may affect mV readouts?

Post by ziemek »

Hi there!

I use a model of Combe et al. 2018: When I removed all tmpo2 variable usage from cell_setup.hoc:
  • Removing this part and all other occurences of tmpo2 variable:

Code: Select all

objref tmpo2
tmpo2=new Shape()
I discovered that in the soma after ~500 ms there are discrepancies in mV readouts:
  • -15.816405568815716 - with Shape (with variable tmpo2)
  • -15.828809384418346 - without Shape (without variable tmpo2)
I did all readouts through Python API, so it may also affect it.

Obviously it may be related to the rest of the code, but I think that tmpo2's Shape is quite independent from the rest of the code.

So does the usage of the Shape may affect mV readouts?
ted
Site Admin
Posts: 5793
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Does the usage of the Shape may affect mV readouts?

Post by ted »

Before commenting I will need to see this for myself. The model authors describe more than one way to launch the model. Exactly what did you do?
ziemek
Posts: 43
Joined: Thu May 23, 2019 8:02 am
Location: Warsaw, Poland
Contact:

Re: Does the usage of the Shape may affect mV readouts?

Post by ziemek »

Dear Ted,

I checked once more the code and I discovered that I use cvode(True), which means that there is a variable timestep.

My current readouts for the same model but different compartment are:

Code: Select all

constant timestep
with shape: ms: 50.00000000000341 mV: -8.280244904579009
no shape:   ms: 50.00000000000341 mV: -8.28024490457921

Variable timestep:
with shape: ms: 50.16482149700094 mV: -8.417091046804176
no shape:   ms: 50.06825794430341 mV: -8.401428051397344
As you can see - for the constant timestep - there is almost identical value of mV. However for the variable timestep - the step differ ~0.1 ms and the readout ~0.016 mV. For this experiment I took the first mV readout which passed 50ms.

So I wonder why the usage of Shape() changes a little bit the timestep?
ted
Site Admin
Posts: 5793
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Does the usage of the Shape may affect mV readouts?

Post by ted »

See my message from June 29.
ted
Site Admin
Posts: 5793
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Does the usage of the Shape may affect mV readouts?

Post by ted »

Did you resolve the problem yourself, or do you still have a question?
ziemek
Posts: 43
Joined: Thu May 23, 2019 8:02 am
Location: Warsaw, Poland
Contact:

Re: Does the usage of the Shape may affect mV readouts?

Post by ziemek »

Sorry Ted, I tried to prepare the code for you and make sure that there is no interference from the core I wrote, but then I switched to a different task, and left it for later, but I will prepare the code next week and let you know.
Post Reply