A reproducible 3D functional manifold on the human cortical organization
Poster No:
1579
Submission Type:
Abstract Submission
Authors:
Xiu-Xia Xing1, Zuo Xi-Nian2
Institutions:
1Department of Applied Mathematics, College of Mathematics, Beijing University of Technology, Beijing, Beijing, 2Beijing Normal University, Beijing, Beijing
First Author:
Xiu-Xia Xing
Department of Applied Mathematics, College of Mathematics, Beijing University of Technology
Beijing, Beijing
Department of Applied Mathematics, College of Mathematics, Beijing University of Technology
Beijing, Beijing
Co-Author:
Introduction:
Human brain function is commonly observed in a very high-dimensional space, which offers supports of the complexity of mind and behavior. System neuroscientists believe the existence of a locally representative low-dimension space across the high-dimensional geometry of the brain function. Such a functional manifold has been explored by using dimension-reduction methods, and named 'connectivity gradients' [1]. Notably, the gradient methods only keep a set of the most strong functional connectivity for consideration (e.g., the top 10% connectivity) [2]. This approaches a tenet to approximate underlying structural connectivity (thus, in principle, a structural manifold) but overlooks the fact that functionally meaningful connectivity are not necessarily high values of functional connectivity as measured by Pearson's correlation between pairs of fMRI time series [3-5]. The functional manifold on the human cortical organization derived from the full fMRI data remains elusive.
Methods:
The data we used in the present study are from the Human Connectome Project (HCP, 1003 healthy young adults) [6] and the Chinese HCP (CHCP, 217 healthy young adults) [7]. All our analyses are based on group average preprocessed whole brain dense functional connectome (.dconn) data identifying temporal correlations between all cortical vertices and subcortical voxels. Specifically, each pre-processed individual data set was temporally trimmed and variance normalization applied and submitted to the principal component analysis (PCA) at the group level. The output of the group-PCA (the top 4500 for HCP or 2500 for CHCP weighted spatial eigenvectors) are then renormalized, eigenvalue-reweighted and correlated to form the group average dense connectome data (91,282 × 91,282 entries). Diffusion map embedding is applied to identify a set of low-dimensional manifolds (i.e., gradients) capturing principal dimensions of spatial variation in connectivity. Notably, this algorithm is implemented in BrainSpace toolbox [2] with the option of no thresholding (normally top 10%) on functional connectivity to derive the functional affinity or similarity of the functional connectivity matrix. The first three gradients span a 3D functional manifold in the human cortical organization.
Results:
The top 3 gradients explain more than half of the observed variances in complete brain connectivity maps (55% for HCP and 52.7% for CHCP). According to the most recent DU15NET [8], the first functional manifold (FM1, Figure 1 top panel) accounts for 27.8% (27.4% for CHCP) variability with one pole in the FPN-B and DN-A / B networks and another pole in the PM-PPr and CG-OP networks, featuring a spectrum from internal to external aspects of the underlying functional processes. The second functional manifold (FM2, Figure 1 middle panel) accounts for 19.0% (16.4% for CHCP) variability with one pole in DN-A, DN-B and language networks, and another pole in FPN-A and SAL/PMN networks, featuring a spectrum from representation to modulation of underlying functional processes. The third functional manifold (FM3, Figure 1 bottom panel) accounts for 8.2% (8.9% for CHCP) variability with one pole in the DN-B SAL/PMN and CG-OP networks, and another pole in the FPN-A and FPN-B networks, featuring a spectrum from self-directed functional processes to goal-directed functional processes.
Conclusions:
The 3D functional manifolds we discovered serve as a framework for the theory of an allostatic-interoceptive system and provide functional characteristics in the cortical organization from the outside to the inside [9,10]. Functional manifolds maintain their promise for both basic neuroscience and clinical translation.
Modeling and Analysis Methods:
Methods Development 1
Neuroinformatics and Data Sharing:
Brain Atlases 2
Keywords:
Data analysis
FUNCTIONAL MRI
Modeling
1|2Indicates the priority used for review
·Figure 1. A 3D functional manifold on the human cortical organization
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
[2] Vos de Wael, R., et al. (2020). BrainSpace: a toolbox for the analysis of macroscale gradients in neuroimaging and connectomics datasets. Communications Biology, 3, 103.
[3] Katsumi, Y., et al. (2023). Correspondence of functional connectivity gradients across human isocortex, cerebellum, and hippocampus. Communications Biology, 6, 401.
[4] Zhang, J., et al. (2020) Intrinsic functional connectivity is organized as three interdependent gradients. Scientific Reports, 9, 15976.
[5] Jiang, C., et al. (2023) Optimizing network neuroscience computation of individual differences in human spontaneous brain activity for test-retest reliability. Network Neuroscience, 7, 1080-1108.
[6] Glasser, M.F., et al. (2016) The Human Connectome Project's neuroimaging approach. Nature Neuroscience, 19, 1175-1187.
[7] Ge, J., et al. (2023) Increasing diversity in connectomics with the Chinese Human Connectome Project. Nature Neuroscience, 26, 163-172.
[8] Du, J., et al. (2024) Organization of the human cerebral cortex estimated within individuals: networks, global topography, and function. Journal of Neurophysiology, 131, 014-1082.
[9] Ibanez, A. and Northoff, G. (2024). Intrinsic timescales and predictive allostatic interoception in brain health and disease. Neuroscience & Biobehaviour Reviews, 157, 105510.
[10] Hilgetag, C.C., et al. (2022). A natural cortical axis connecting the outside and inside of the human brain. Network Neuroscience, 6, 950-959.