Brain Hierarchical Atlas 2 (BHA2)

Laburpena

Elucidating the intricate relationship between the structure and function of the brain, both in healthy and pathological conditions, is a key challenge for modern neuroscience. Magnetic Resonance Imaging (MRI) has helped in the understanding of this matter, with diffusion images providing information about structural connectivity (SC) and resting-state functional MRI revealing the functional connectivity (FC). Furthermore, the brain operates by discrete multiscale computations in both the time and spatial domains, in a way that is far from known (Churchland and Sejnowski, The MIT Press, 1994). To advance in the understanding of this puzzle, a dual structure-function hierarchical clustering strategy was proposed in (Diez et. al, SciRep, 2015), providing a common skeleton shared by structure and function. Here, we further extend this approach by:<br> 1. Fine-tuning the amount of matching between SC and FC via a free-parameter gamma. Specifically, when gamma is set to 0, SC is fully recovered, while when gamma is set to 1, FC is recovered. In between these extremes, a fusion scenario occurs, where both SC and FC contribute to the connectivity patterns. The raw data to generate the SC and FC matrices came from (Babayan et. al, Scientific Data, 2019), and can be downloaded from https://fcon_1000.projects.nitrc.org/indi/retro/MPI_LEMON.html.<br> 2. Making use of brain-transcriptomic data to shed light on biological interpretability of brain-related diseases in the gamma-modulated multiscale structure-function correspondence.<br> 3. Providing to the scientific community open data of different scenarios of structure-function sharing and at different spatial scales, and open code to generate them in a MRI dataset. The dataset is organized in the following way: data<br> │ ├───iPA_nROIS [Different spatial scales 183, 391, 568, 729, 964, 1242, 1584, 1795 and 2165]<br> │ │ ├───iPA_nROIS.nii.gz [Brain parcellation image]<br> │ │ ├───iPA_nROIS.csv [MNI Coordinates and location of the brain parcellation ROIs]<br> │ │ ├───SC [Structural connectivity matrices]<br> │ │ ├───FC [Functional connectivity matrices]<br> │ │ ├───ts [Resting-state functional connectivity timeseries]<br> │ │ | ├───confounds [Confounds used to filter the timeseries]<br> │ │ ├───gamma-trees [Gamma-trees of nROIs levels]<br> │ │ ├───transcriptomics.csv [Transcriptomics' of each ROI] If you want to use this dataset, please cite: <em>Antonio Jimenez-Marin, Ibai Diez, Asier Erramuzpe, Sebastiano Stramaglia, Paolo Bonifazi, Jesus M Cortes</em>. <strong>Open datasets and code for multi-scale relations on structure, function and neuro-genetics in the human brain</strong>. biorxiv. 2023. https://doi.org/10.1101/2023.08.04.551953