search

openworm / OpenWorm

Repository for the main Dockerfile with the OpenWorm software stack and project-wide issues
http://openworm.org
MIT License
2.67k stars 208 forks source link

Produce muscle cell electrophysiology trace figure and compare with simulated match #237

Open slarson opened 7 years ago

slarson commented 7 years ago

Would show activity of the muscle cell and compare to real data. Two papers in the muscle model repo have data for comparison that can be used to produce this figure.

slarson commented 7 years ago

Pre-reqs to accomplishing this are: https://github.com/openworm/muscle_model/issues/64 and https://github.com/openworm/muscle_model/issues/38

slarson commented 7 years ago

Discussion today led to updating https://github.com/openworm/muscle_model/issues/64 and producing a check list. We'll be working on completing that checklist now.

slarson commented 7 years ago

Talked to @rgerkin about this last week! Any updates yet @rgerkin ?

rgerkin commented 7 years ago

@slarson I now have it running successfully on my machine, and in Python 3.

A question about biological context: in the data figure they show the spontaneous activity of the muscle cell, in the absence of any particular stimulation. It appears that the simulation we are running includes both a neuron and a muscle cell, with the neuron doing basically nothing and the muscle cell exhibiting spiking. Are we operating under the assumption that it is not spontaneous activity in the neuron driving the muscle cell (in the real worm), but rather intrinsic mechanisms causing the activity of the muscle cell? If it is the former, it seems like we need the neuron to do something. If it is the latter, does the cell even need to be connected in this model for the purposes of testing the muscle cell?

slarson commented 7 years ago

Are we operating under the assumption that it is not spontaneous activity in the neuron driving the muscle cell (in the real worm), but rather intrinsic mechanisms causing the activity of the muscle cell? If it is the former, it seems like we need the neuron to do something. If it is the latter, does the cell even need to be connected in this model for the purposes of testing the muscle cell?

In this figure, the muscle activity could be caused by a combination of intrinsic mechanisms in the muscle and potentially some background synaptic input from the neuron.

If we want to isolate the muscle, the right-most trace in 2A may actually be more appropriate, as they have blocked the neurotransmitter receptors...

In terms of the presence of the neuron in the simulation -- if it is inactive it should have no effect. Its presence right now is I believe just a side effect of the c302 framework.

rgerkin commented 7 years ago

OpenWorm integration