This project contains the preregistration, data and analyses scripts of the the project "The functional neurobiology of negative affective traits across regions, networks, signatures, and a machine learning multiverse"
The Patterns predicting vulnerability to stress from the Training and the full data (Training + hold-out) can be found in "Results\holdOutModels" as neoN6_IAPS_LookNeg-vs-LookNeut_train_GM_cente_cv_svr.nii and neoN6_IAPS_LookNeg-vs-LookNeut_full_GM_cente_cv_svr.nii
The main Analysis scripts are contained in the top Level Folder. To run them, you need to download the canlab coretools Folder and the canlab_single_trials Folder from GitHub and put them on your matlab path. Also, you need to put the subfolder "Functions" on your path, which contains some custom functions for the present study.
The neuroimaging data will be shared via a neurovault link in this description upon the acceptance of the final paper. These files have to be donwloaded into a folder named "Subject-level-maps" within the Data-folder already existing in the current repository. If there is a published peer-reviewed paper, but you do not see a link here for some reason, please contact: maurizio.sicorello@zi-mannheim.de
A brief description of the analyses scripts can be found on top of the respective script. In most cases, relative paths to the script location were used.