calculate_network_bursts

[sofiapuvogelvittini]

Dear meaRtools authors, thank you for this great package.
I am trying the calculate_network_bursts on my s object.
Looking inside the functions that are called inside calculate_network_bursts, I can see that there are some parameters that may vary depending on the samplig rate. Could you indicate me please wich would be the best way to modify this and set the function according to my expriment?

Thank you very very much for yout time.

[sje30]

Hi,

thank you for submitting this, and sorry for the delay in replying.

Could you explain in a bit more detail what you are trying to do?

[sofiapuvogelvittini]

Dear sje30, thanks for your response.
I have data from two different systems, axion and multichanell systems multiwell MEA. We want to start with the timestamp data from each system so we can analyze these in the same way. Axion has 12500 hz of sampling rate, while MCS has 10 000 hz. Do you think I should modify some parameter for burst or NB detection to work better with the two sampling rates?
My second question: I am not sure how I can interpretate the variable "mean_NB_time_per_sec". Looking at your paper, it says that it corresponds to the aaverage NB length (Table 4), however the values that I obtain are much more smaller than for electrode burst (not NB burst). Is the mean_NB_time_per_sec the duration in sec or ms?
Third question: Can I obtain the average inter spike interval within NB? would be this the same as 1/spike_intensity?
Thank you!

[sofiapuvogelvittini]

Also, I don't know how to interpretate the variables related to network bursts in table S4. For all of them, it is indicated [windo size in ms]. Does this mean that variables are in ms? is spike intensity in ms too?

[sje30]

Thanks for your questions. I'm afraid that I believe these network burst algorithms were quite new, compared to the other routines (that i am more familiar with), and probably were only tested on Axion.

The sampling frequency of 12500 Hz does indeed seem to be hard coded into meaR:::.nb_extract_features

As to your other questions, I'm afraid I don't know, but I'll look into them. I hope to resume work on this package this year, as several issues have now accumulated. It may take a few months though, I'm afraid