Last thing: Please watch @eliobalestrieri.eurosky.social for a 🧵on the computational side - simulations, notebooks, and tickling the HPC cluster with 20x 1093680 parameter scenarios.

It was so much fun drawing this up on a flip chart with @ualsbombe.bsky.social over too many cups of espresso. Incredibly proud of Teresa and Elio for what became of these ideas. Not least, thanks to @martinasaltafossi.bsky.social and Laura Bock Paulsen for rigorous testing and all the discussions.

Finally, in the tradition of 'Ten simple rules ...' papers, we boil our tutorial down to the key aspects to look out for. From conceptual decisions to practical advice, we hope this helps you avoid some of the mistakes we have made. Learning from them is how this paper came to be.

In addition to this benchmarking, we quantify the statistical power of circular cluster permutation testing as a function of the number of trials and participants, granularity of phase binning, and the width of the moving window used. This lets you assess what to expect given your own parameters.

Our cluster permutation approach for circular data is not at all limited to respiratory time series and comes as an out-of-the-box usable tool for your own analyses. We provide detailed analyses of type I/II error control, plus example data and a full executable pipeline. Oh, the simulations...

Struggles with phase-amplitude coupling analyses often come down to constructing an adequate null distribution to compare your empirical findings against. We show how strongly your inference will depend on your null, when and why shuffling is an option, and suggest IAAFT as a general approach.

Based on years of head-scratching and running into ever-new issues with previous methods approaches, we systematically compare the most important analysis parameters and offer guidance on best-practices choices. Again, everything available in #python and #matlab code.

In a step-by-step tutorial, we highlight critical choices and potential pitfalls as we go from collecting raw data all the way to the final analysis. With all code openly accessible, this may serve as a blueprint for your own studies on respiration-brain coupling.

📂 github.com/teresaberthe...

Congratulations Mara - looking forward to everything coming out of the amazing research programme you have planned for Berlin!

What a pleasure to have you here - thanks again and until next time!