Problems understanding how to generate firingRates array (MATLAB)

[maristanye]

Dear Dmitry,

thanks for the code and the elife paper on dPCA. I am trying to get started with it for my data set, however I struggle in understanding how to generate the firingRates matrix (despite looking through the demo)

Lets say I have:

N= 30 neurons;
T= A trial length of 60 data points
maxTrialNum= 135.

So e.g. this is represented as a 3D 30x60x135 matrix already in my data structure.

And then I have two different conditions (S) and two different Decisions (D) in my behaviour which are represented as two different 1D 135x1 arrays (with 0 and 1 for right/left stimuli/decisions).

I struggle wrapping my head around how to produce from this your

firingRates: N x S x D x T x maxTrialNum

data structure to get started with the dPCA (should it end up being a 30x2x2x60x135 array?). I guess my head stops working at 3 dimensions.
Anyhow thanks a lot for the tool and have a nice weekend,
Eduardo

[dkobak]

Dear Eduardo,

it's been a while since I worked with Matlab... In Python this would look something like this:

import numpy as np

X = np.random.randn(30, 60, 135)
S = np.random.randint(2, size=135)
D = np.random.randint(2, size=135)

firingRates = np.zeros((30, 2, 2, 60, 135)) * np.nan

maxTrialNum = 0

for s in (0, 1):
    for d in (0, 1):
        trials = (S == s) & (D == d)
        firingRates[:, s, t, :, :np.sum(trials)] = X[:, :, trials]
        
        if np.sum(trials) > maxTrialNum:
            maxTrialNum = np.sum(trials)
            
firingRates = firingRates[:, :, :, :, :maxTrialNum]

Does this help?

[maristanye]

Dear dkobak,

thanks a lot for the help and sorry for the late reply. Yes that was very helpful and I was able to run the analysis. I have one more question. Can I combinedata from different imaging sessions from different neurons in this analysis? An example:

Day1:
N= 42 neurons;
T= A trial length of 60 data points
Trials= 48.
->42x60x48
Day2:
N= 28 neurons;
T= A trial length of 60 data points
Trials= 52.
->28x60x52

would the resulting firingRates be a 70x2x2x60xmaxTrialNum where some of the values in the 1D (70 neurons =42neurons[day1]+28neurons[day2]) are NaNs depending if it is data from the corresponding session or not?

Cheers,
Eduardo

[dkobak]

Yes you can combine data from different sessions. All datasets analyzed in our eLife paper were constructed like that (from recordings obtained over many sessions).

You have two arrays, 42 x 60 x 48 and 28 x 60 x 52. You can rearrange both as I explained earlier, so they will become 42 x 2 x 2 x 60 x maxTrialNum1 and 28 x 2 x 2 x 60 x maxTrialNum2. To combine, find maxTrialNum = max(maxTrialNum1, maxTrialNum2) and pad one of the arrays with nans along the 5th dimension so that the arrays become 42 x 2 x 2 x 60 x maxTrialNum and 28 x 2 x 2 x 60 x maxTrialNum. Now you can concatenate along the 1st dimension, to get 70 x 2 x 2 x 60 x maxTrialNum.

[maristanye]

yes, thanks, thats what I just figured out. I wrote a MATLAB function to do these steps and generate firingRates and trialNum. Let me know if you are interested.

Thanks a lot again,
Eduardo