Causal Cortical and Thalamic Connections in the Human Brain

Causal Cortical and Thalamic Connections in the Human Brain preprint This repository contains research data to support a scientific publication. Here you can find the example codes and preprocessed, trial-averaged data to replicate the reported findings. Abstract: The brain's functional architecture is shaped by causal connections between its cortical and subcortical structures. Here, utilizing single-pulse electrical stimulation in 27 participants implanted with 4864 intracranial electrodes, we mapped the causal connections throughout the human brain, encompassing both cortical and subcortical regions including multiple thalamic nuclei. In particular, we dissociated three unique spectral patterns generated by the perturbation of a given brain area. Among these, a novel waveform emerged, marked by distinct temporal and spatial patterns uniquely associated with the thalamus. Stimulations of the thalamus caused delayed-onset theta oscillations in both ipsilateral and contralateral cortices. Moreover, features of causal connectivity in the human brain, as identified here, revealed highly organized patterns which are different from each other, suggesting that each feature is related to a distinct signal transmission pathway. Our causal connectivity data can be used to inform biologically informed computational models of the functional architecture of the brain. Data type: intracranial electrophysiological recording upon repeated trials of single-pulse electrical stimulation. Sample size: Twenty-seven participants (40.7% female) diagnosed with focal epilepsy were recruited in this study. The total number of SEEG electrode contacts being studied was 4864. Each participant had 180 ± 46 (mean ± standard deviation) contacts implanted, 6.8+/-4.2 bipolar channels implanted in the thalamus, with at least two subdivisions of the thalamus sampled (2.3+/-1.8 bipolar channels per thalamic subdivision per person). Single-pulse electrical stimulation was repeated 44+/-2 times to establish a connection. The initial number of stimulated-recorded pairs across the whole-brain regions to be studied was 275336, from which 76899 pairs were determined to be connected and proceeded with the subsequent analyses. For the connected pairs, the number of pairs with stimulation channels in the thalamus was 7566 (291.0+/- 250.0 per subject; 109.0+/-78.7 for antTH, 72.5+/-118.7 for midTH and 109.5+/-108.5 for pstTH). The whole brain areas were divided into 26 regions of interest in our study, there were 14.3+/-3.8 regions covered with electrodes per person. When covered, the total number of pairs with stimulation channels in the regions other than the thalamus was 206.2+/-233.1 per subject per region. Used softwares and toolboxes: MATLAB>=R2021a PYTHON == ==3.9 R ==== 4.2 umap-learn == ==0.5.3 mne ==== 1.5.0 r-nlme == ==3.1_162 jupyter-matlab-proxy ==== 0.7.1 Folder organization: - COHORT: sample size information in the cohort- DATA: repeated single pulse evoked responses (RSEPS) in time and time-frequency domains; - metaTable.csv (contains information about all examined pairs of connections) - if the data structure contains a filteridx (Filter Index) field, it corresponds to the meta-table row order. By default, the data is organized in the same order as the meta-table. - RSEPS - time_domain: stimulation evoked potentials at the time domain (trial-averaged), sampling rate = 1000. The epoched range is (-200, 1000]ms around a stimulation. - timeFreq_domain: stimulation evoked spectrograms (power and inter-trial phase coherence), sampling rate = 1000 - UMAP - input data for machine learning- scripts: customized codes for producing the results- results: - umap: - semi-supervised learning: activation detection results for each individual - other files are the outcome of the group-level UAMP learning - feature: - feature curves for F1 (clst0), F2 (clst1) and F3 (clst2) - stats: - clstPermutation: cluster-based permutation results as reported in the paper - ROIanalysis: comparing feature representation among thalamic nuclei and between pathways, as reported in the paper - **metaTable variable explained**: 1. **subject, stim_chan, record_chan**: subject ID, bipolar stimulating/recording channel labels 2. **JP_label_in1, JP_label_in2, JP_label_out1, JP_label_out2**: Region name by Josef's manual labelling. The suffix "out"/"in" refers to stimulating and recording channels respectively, i.e. signal sending OUT and going IN. The suffix 1/2 refers to the order of the bipolar pair, i.e. for A1-A2, JP_label_in1 would be referring to the A1. 3. **Yeo7_out1, Yeo7_out2, Yeo7_in1, Yeo7_in2**: Intrinsic functional connectivity network labelling for the referred channel. Network parcellation follows Yeo-7 network from the JoN publication 2011. 4. **MNIout_coord_1, MNIout_coord_2, MNIout_coord_3**: MNI coordinates for the stimulation channel. The MNI coordinates take the middle point of the bipolar pair. Suffix 1, 2,3 corresponds to the x, y, z axis in the MNI space. 5. **MNIin_coord_1, MNIin_coord_2, MNIin_coord_3**: MNI coordinates for the recording channel. 6. **FSout_coord_1, FSout_coord_2, FSout_coord_3**: Coordinates of the stimulation channels in the FS_LR surface space. FS_LR is a standard suaface space with homologous left and right hemispheres. These coordinates have been projected to the right hemisphere, useful for certain visualization purpose. One needs to use a FS_LR standard surface for correct visualization. 7. **FSin_coord_1, FSin_coord_2, FSin_coord_3**: Coordinates of the recording channels in the FS_LR surface space. 8. **umapAct**: binary vector indicating whether an "activation" is labelled by the UMAP semi-supervised learning algorithm. 0 = non-activated, 1 = activated, NA = noisy data. 9. **peak_maxCor_clst1, peak_maxCor_clst2, peak_maxCor_clst3**: The peak value of the feature representations (R) acorss time. Suffix clst1/2/3 corresponds to feature 1/2/3. 10. **time_maxCor_clst1, time_maxCor_clst2, time_maxCor_clst3**: The time delay to the peak R upon stimulation.