limbiccortical

This is a repostitory for the project "limbic-cortical", where we applied the limbic-cortical model of major depression to healthy subjects with particular risks for depression.

The final paper as well as supplementary data can be found here: https://www.frontiersin.org/articles/10.3389/fnsys.2020.00028/full

This repository contains analysis scripts, as well as intermediate data.

Data in this repository

fMRI group level analysis

statistical analysis (e.g. SPM file, contrast image, t-map)
- navigate to 'data/second_level_analysis/' The 'SPM.mat' file should be accessible via MATLABs SPM toolbox, MATLAB itself, Python (scipy.io access), ... The t-map or contrast image is accessible with each software that handles '.nii' images.

fMRI first level analysis

statistical contrast images & t-maps of the single subjects
navigate to 'data/first_level_analysis' There you can find the contrast images of the [faces>houses] contrast of each subject (subject ID is in the file name).

DCMs

navigate to '/data/dcms/' You find the estimated DCM models (12 per subject) sorted into corresponding risk groups:
- no_risk
- gen_risk ("genetic risk", family history or genetic liability)
- env_risk ("environmental risk", childhood maltreatment)
- both_risks

Each subject has 12 models. The subject ID is in the filename (e.g. 0123). Each of the 12 models has a different structure (differing in B- and C-matrices). You can read the structure in the file name as follows:

Ainp0 --> Amygdala Input = 0 (C-matrix, not existent)
Ainp1 --> Amygdala Input = 1 (C-matrix, existent)
Oinp0 --> mPFC (O or ORB because of "Orbitofrontal") input = 0 (C-Matrix, not existent) ...
AO0 --> connection from Amygdala to mPFC is not modulated by regressor "faces" (B-matrix)
AO1 --> connection from Amygdala to mPFC is modulated by regressor "faces" (B-matrix) ...

Analyses in the repostitory

Bayesian Estimation (BEST)

Pipeline located in the folder: rk_BEST_analysis.R

The BEST library needed is located in folder "packages".

Multiple Linear Regression analysis (see supplementary methods)

R scripts located in the folder "linearModel"

other small scripts

rk_get_oo_and_xp.m Matlab script to get the posterior model probabilities as well as exceedance probabilites of each group (need BMS.mat for that)

rk_group_characteristics.R Just outputs some group characteristics (e.g. age a.s.o. for description in the article).

kesslerr/limbiccortical