[Camino-users] Unexpected Monte Carlo Diffusion Simulations

Mon Sep 22 23:43:43 PDT 2014

Hi Jonathan,

Thanks for your reply. I used “intracellular volume fraction” = “icvf”.

Here is a list of all the substrate geometric properties used for the data in the graphic:

radius, r

axonal spacing, a

diameter

a/r

intracellular volume fraction, icvf

5.967E-08

5.082E-07

1.193E-07

8.52

0.05

3.375E-07

2.875E-06

6.750E-07

8.52

0.05

1.350E-06

1.150E-05

2.700E-06

8.52

0.05

5.400E-06

4.600E-05

1.080E-05

8.52

0.05

3.055E-05

2.602E-04

6.110E-05

8.52

0.05

4.505E-07

1.108E-06

9.009E-07

2.46

0.6

9.009E-07

2.215E-06

1.802E-06

2.46

0.6

1.802E-06

4.431E-06

3.604E-06

2.46

0.6

3.604E-06

8.861E-06

7.208E-06

2.46

0.6

7.208E-06

1.772E-05

1.442E-05

2.46

0.6

1.442E-05

3.544E-05

2.883E-05

2.46

0.6

2.883E-05

7.089E-05

5.766E-05

2.46

0.6

5.766E-05

1.418E-04

1.153E-04

2.46

0.6

1.153E-04

2.836E-04

2.306E-04

2.46

0.6

3.185E-07

7.832E-07

6.371E-07

2.46

0.6

6.371E-07

1.566E-06

1.274E-06

2.46

0.6

1.274E-06

3.133E-06

2.548E-06

2.46

0.6

2.548E-06

6.266E-06

5.096E-06

2.46

0.6

5.096E-06

1.253E-05

1.019E-05

2.46

0.6

1.019E-05

2.506E-05

2.039E-05

2.46

0.6

2.039E-05

5.013E-05

4.077E-05

2.46

0.6

4.077E-05

1.003E-04

8.154E-05

2.46

0.6

8.154E-05

2.005E-04

1.631E-04

2.46

0.6

1.631E-04

4.010E-04

3.262E-04

2.46

0.6

3.502E-06

7.235E-06

7.004E-06

2.07

0.85

7.004E-06

1.447E-05

1.401E-05

2.07

0.85

1.401E-05

2.894E-05

2.802E-05

2.07

0.85

2.802E-05

5.788E-05

5.604E-05

2.07

0.85

5.604E-05

1.158E-04

1.121E-04

2.07

0.85

1.121E-04

2.315E-04

2.241E-04

2.07

0.85

6.191E-07

1.279E-06

1.238E-06

2.07

0.85

3.962E-05

8.186E-05

7.925E-05

2.07

0.85

3.170E-04

6.548E-04

6.340E-04

2.07

0.85

Thanks,
David

De : jonhrafe at gmail.com [mailto:jonhrafe at gmail.com] De la part de Jonathan Rafael Patino
Envoyé : 18 September 2014 21:18
À : Slater David
Cc : camino-users at www.nitrc.org
Objet : Re: [Camino-users] Unexpected Monte Carlo Diffusion Simulations

Hi David.

I don't  really have an explanation for the weird behavior of the radial diffusivity. Can you tell us the set of radius and separation values that you used in your experiments? I would like to visualize the trajectories of the spins in each simulation.

Another thing, in your graphic, what's the meaning of "icvf"? I'm not familiar with the concept.

Best!

2014

-09-16 10:34 GMT-05:00 Slater David <David.Slater at chuv.ch<mailto:David.Slater at chuv.ch>>:
Hi everyone,

I’ve been working with the Camino Monte Carlo diffusion simulator. To check things were working correctly I tried some simple tests of diffusion around hexagonally packed cylinders. Spins were placed only in the extracellular space by using the ‘extra’ option. However even with these simple tests I’ve got some unexpected results.

I created a number of substrates with different axon radii while keeping the ratio of cylinder separation to axon radius constant. In this way we scale up and down the characteristic length scales of the system while keeping the intra- and extra-cellular volume fractions constant.

By applying the same scheme to each substrate with (Δ-δ/3)=30ms we would expect radial diffusivity to approach free diffusivity in systems with very large length scales (i.e. relatively few membrane interactions) while radial diffusivity should approximate theoretical tortuosity model predictions in systems with very small characteristic length scales.

In fact what we find from the simulations is that radial diffusivity can go up as the axon radius and axonal spacing go down. See the attached figure. This seems physically implausible to me: how is it that extracellular diffusivity can increase in systems with more membranes per unit length?

Here are example commands that I used to create the .traj files and to generate the synthetic measurements:

datasynth -walkers 200000 -tmax 5000 -p 0 -voxels 1 -initial extra -duration 0.04 -diffusivity 17E-10 -substrate cylinder -packing hex -cylinderrad 9.01E-07 -cylindersep 2.22E-06 -trajfile ic60_diam1p8E-06_extra_hex_dura40ms.traj

scan -trajfile ic60_diam1p8E-06_extra_hex_dura40ms.traj -schemefile J30_DELTA30.scheme2 | float2txt > ic60_diam1p8E-06_extra_hex_dura40ms_J30_DELTA30.txt

I have also attached the .scheme2 file that was used.

What might be leading to these unexpected results? Do you have any suggestions for where I might be going wrong?

Any help or suggestions will be greatly appreciated.

Thanks,
David

_______________________________________________
Camino-users mailing list
Camino-users at www.nitrc.org<mailto:Camino-users at www.nitrc.org>
http://www.nitrc.org/mailman/listinfo/camino-users

J

.

 R.

[https://t.yesware.com/t/9d7353696357d0e6d056edc2a85a7e16f61c007f/3a89ec9bf42d264e00269c67cfe2b752/spacer.gif][http://t.yesware.com/t/9d7353696357d0e6d056edc2a85a7e16f61c007f/3a89ec9bf42d264e00269c67cfe2b752/spacer.gif]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/camino-users/attachments/20140923/2178a060/attachment-0001.html>