NITRC cPPI Toolbox for fMRI Forum: open-discussion

mostly positive values

Adam Kaminski — Mon, 24 Oct 2022 16:16:45 GMT

Hello,

Thank you for the wonderful toolbox!

I'm specifying a contrast with two conditions and get group averaged matrices which are mostly positive (range = -0.04 to 0.87). This just struck me since I had been expecting covariations in task-related activity to be more evenly positive and negative. Is this not the right expectation or is there something I may be doing wrong?

Many thanks!
Adam

RE: contrast

Selma Lugtmeijer — Tue, 14 Jun 2022 21:28:51 GMT

Thank you for your explanation!

RE: contrast

Alex Fornito — Fri, 10 Jun 2022 21:25:33 GMT

[i]Originally posted by Selma Lugtmeijer:[/i][quote]Hi,

thank you for making this toolbox available! I have two questions:

Does there always need to be a contrast of 2 conditions? What if I want to see in 1 condition if 2 groups of subjects differ in strength of functional connectivity for certain ROI pairs?

And a more theoretical question, is creating a FC matrix for task condition 1 and 2 separately and subtracting them, the same as creating one FC matrix for the contrast 1-2?

Thanks,
Selma[/quote]
[color=#000000]Hi,[/color]

[color=#000000]You could in principle use a [1 0] contrast to map FC in condition A and then [0 1] for condition B and then compare the two, but it will change the interpretation. In this case, the FC matrix you map for each condition will be condition A vs the implicit baseline (i.e, everything else), which is slightly different to a [1 -1] scenario, where the FC matrix captures coupling between the task-modulated component of regional signals.[/color]

[color=#000000]Ultimately it depends on whether you think the [1 0] scenario is interpretable.[/color]

[color=#000000]hth[/color]
[color=#000000]A [/color]

contrast

Selma Lugtmeijer — Fri, 10 Jun 2022 12:05:21 GMT

Hi,

thank you for making this toolbox available! I have two questions:

Does there always need to be a contrast of 2 conditions? What if I want to see in 1 condition if 2 groups of subjects differ in strength of functional connectivity for certain ROI pairs?

And a more theoretical question, is creating a FC matrix for task condition 1 and 2 separately and subtracting them, the same as creating one FC matrix for the contrast 1-2?

Thanks,
Selma

RE: time series extraction for cPPI

Alex Fornito — Thu, 10 Mar 2022 7:56:26 GMT

[i]Originally posted by noamagal:[/i][quote]Dear Dr. Fornito,
thank you for devloping this wonderful toolbox.
so far, i've already performed the first level for all my subjects in SPM12, so i have a SPM.mat file for each subject.
i also have my ROIs as .mat files (created using MARSBAR).
i ran into a technical question when trying to preform cPPI analyses on my data, hopes it's ok -
in the "cPPI_master.m" code lines 93-103 the user is expected to enter the ROI time courses.
i'm not sure how to extract these time courses at this point using SPM.
when i run gPPI models using the gPPI toolbox this step is already implemented as a part of the PPPI.m function, so i'm not sure how to do this manually.
hope you can help me in this issue.

many thanks,
Noa Magal.[/quote]
[color=#000000]Hi Noa[/color]
[color=#000000]You should be able to extract the ROI time series using Marsbar as per instructions (http://marsbar.sourceforge.net/marsbar.pdf).[/color]
you would then need to reformat as a cell array, as outlined in cPPI_master.m
hth
A

time series extraction for cPPI

noamagal — Thu, 10 Mar 2022 6:57:27 GMT

Dear Dr. Fornito,
thank you for devloping this wonderful toolbox.
so far, i've already performed the first level for all my subjects in SPM12, so i have a SPM.mat file for each subject.
i also have my ROIs as .mat files (created using MARSBAR).
i ran into a technical question when trying to preform cPPI analyses on my data, hopes it's ok -
in the "cPPI_master.m" code lines 93-103 the user is expected to enter the ROI time courses.
i'm not sure how to extract these time courses at this point using SPM.
when i run gPPI models using the gPPI toolbox this step is already implemented as a part of the PPPI.m function, so i'm not sure how to do this manually.
hope you can help me in this issue.

many thanks,
Noa Magal.

RE: cPPI vs gPPI

Alex Fornito — Wed, 18 Feb 2015 22:04:35 GMT

Dear David,
Sorry for the delay (I did not see this post).

There are a few potential reasons for this discrepancy. Perhaps the biggest difference is the model. The model in gPPI includes PPI terms for all relevant task regressors, whereas cPPI only includes the PPI term for the contrast of interest only. To allow a more equivalent comparison, you would probably need to tweak cPPI to use a model more similar to gPPI.

Obviously, the directionality is also an issue, so you would need to see what the correlations are going in both directions.

Regards
Alex

[i]Originally posted by David Coynel:[/i][quote]Dear Dr Fornito, thanks for sharing this nice connectivity tool. I have a general question about the comparability of your approach with the standard PPI or the generalized PPI (http://www.nitrc.org/projects/gppi/). I understand that with the latter you create an asymmetric connectivity matrix, which is not always easy to interpret in terms of functional connectivity.

I ran both a cPPI and a gPPI in several subjects for a set of seeds, and was expecting to get connectivity values that were well correlated between the two methods. However what I observe is that within a given subject the values are rather anticorrelated, meaning that high cPPI correlation values are usually associated with lower gPPI beta estimates.

I ran both analyses using the standard scripts, so I hope I did not do anything wrong, and I would be glad to hear your opinion on that.

David Coynel[/quote]

cPPI vs gPPI

David Coynel — Tue, 21 Oct 2014 15:18:22 GMT

Dear Dr Fornito, thanks for sharing this nice connectivity tool. I have a general question about the comparability of your approach with the standard PPI or the generalized PPI (http://www.nitrc.org/projects/gppi/). I understand that with the latter you create an asymmetric connectivity matrix, which is not always easy to interpret in terms of functional connectivity.

I ran both a cPPI and a gPPI in several subjects for a set of seeds, and was expecting to get connectivity values that were well correlated between the two methods. However what I observe is that within a given subject the values are rather anticorrelated, meaning that high cPPI correlation values are usually associated with lower gPPI beta estimates.

I ran both analyses using the standard scripts, so I hope I did not do anything wrong, and I would be glad to hear your opinion on that.

David Coynel

Welcome to Open-Discussion

Alex Fornito — Tue, 27 Aug 2013 2:20:39 GMT

Welcome to Open-Discussion