Hi Sim,

All of these analyses can indeed be performed in FSLnets, which is a Matlab toolbox. You need to download the toolbox as explained on the website (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLNets), and then start working with nets_examples as a guide.

Estimating partial correlation can be done using the ‘ridgep’ option in nets_netmats, and a one-group t-test can be performed with the nets_glm script. What we did for the paper was average each element across the two connectivity matrices (one for shock and one for safe) within subject first, before entering the resulting averaged matrix into the t-test.

Best wishes,

Janine

-----
Dr Janine Bijsterbosch
Postdoctoral Researcher
FMRIB Centre, University of Oxford
John Radcliffe Hospital
Oxford, United Kingdom
[log in to unmask]ac.uk

On 30 Jan 2017, at 18:35, Sim Luck <[log in to unmask]> wrote:

Hi, Anderson,

Thanks so much for your kind message. I asked that question as I am trying to do a network analysis, which is based on basic ROI analysis, perhaps. 

In detail, Bijsterbosch, Smith & Bishop. (2015, JOCN) wrote something below (page 1845, link: http://bishoplab.berkeley.edu/Bijsterbosch_et_al_JCognitiveNeuroscience_2015.pdf): 

The mean of the two connectivity matrices estimated for “safe” and “shock anticipation” periods was calculated for each participant. A two-tailed one-group t test was performed using the resulting mean connectivity matrix for each participant to determine group-level connectivity between the four ROIs. Results are summarized in Figure 1". In their study, they chose 4 ROIs. I am puzzled how they did that analysis.  In page 1844, the authors did say "Partial correlation with Tikhonov regularization (0.1), as provided by FSLnets (fsl.fmrib.ox.ac.uk/fsl/fslwiki/ FSLNets), was used to calculate 4 × 4 connectivity matrices for each participant and for each scan". I did not understand this very well.

I was thinking that they must ONLY have one connectivity matrix, NOT two matrices, as the signal change should be "Shock anticipation" vs. "Safe" (i.e.,. safe is the baseline). Am I correct? One method perhaps is that we  first extract the BOLD signal changes for each ROI, then do the Perason Correlation, using Matlab. Does the FSLnets help? Anyone can advise me how we can use this FSLnets?

The next question is how we can do the two-tailed one-group t test using the resulting mean connectivity matrix for each participant, and determine group-level connectivity? What should be the input for the t-test?

Thank you very much,

Sim Luck

On Sun, Jan 29, 2017 at 7:12 PM, Anderson M. Winkler <[log in to unmask]> wrote:

Hi Sim Luck,

What do you mean by ROI analysis?

The best tip for FSLnets is the manual: https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLNets

All the best,

Anderson

On 26 January 2017 at 19:19, Sim Luck <[log in to unmask]> wrote:

Hi, List experts,

I am trying to do the ROI analysis.  Anyone can introduce me something about this with FSL?

I am also thinking to use FSLnet for network analysis after I get the ROI signal. Anyone can advise me on the technique tips?

Thank you very much,

Sim Luck