Poster No:
1603
Submission Type:
Abstract Submission
Authors:
Sabrina Primus1, Felix Hoffstaedter2, Federico Raimondo2, Juliane Winkelmann1, Simon Eickhoff2, Johannes Müller3, Kaustubh Patil2, Konrad Oexle1
Institutions:
1Institute of Neurogenomics, Helmholtz Munich, Munich, Bavaria, 2Research Center Jülich, Jülich, NRW, 3Center for Mathematics, Technische Universität München, Munich, Bavaria
First Author:
Sabrina Primus
Institute of Neurogenomics, Helmholtz Munich
Munich, Bavaria
Co-Author(s):
Johannes Müller
Center for Mathematics, Technische Universität München
Munich, Bavaria
Konrad Oexle
Institute of Neurogenomics, Helmholtz Munich
Munich, Bavaria
Introduction:
The morphology of the human brain provides information about physiological and pathogenic processes such as neurodevelopment, aging, and neurodegeneration. Previous genetic studies set the focus on more global univariate observables such as brain structure volume and surface area which were shown to be highly heritable (Hibar,2015). However, that approach cannot discriminate between shapes for which those observables are equal while they differ locally in form. A valid shape descriptor is given by the spectrum (set of eigenvalues) of the Laplace-Beltrami operator (LBS) acting on the object's surface. It provides a multivariate representation of shape, purely based on intrinsic geometric properties (Reuter,2006). Our previous multivariate genome-wide association study (GWAS) on the LBS of 22 subcortical brain structures in 19862 White British UK Biobank participants revealed 80 unique genomic variants affecting shape and known associations with neurodegeneration and mental disorders (Primus,2024). The biological interpretation of these variants is complicated since post-GWAS tools are often designed to handle only univariate and not multivariate data. Therefore, we investigated dimensional reduction of the LBS using principal component analysis (PCA) in order to obtain a more compatible univariate proxy of shape.
Methods:
We used quality controlled imputed genotype data of 19862 healthy unrelated White-British individuals (age range: [46.0,81.7] with mean±SD of 64.3±7.4) and their structural T1w MRI data. We applied FreeSurfer and the BrainPrint Python package (Wachinger,2015) to compute the LBS on triangular meshes for 22 subcortical structures: 4th ventricle, brain stem, accumbens area (left and right), amygdala (l+r), caudate (l+r), cerebellum cortex (l+r), cerebellum white matter (l+r), hippocampus (l+r), pallidum (l+r), putamen (l+r), thalamus proper (l+r), and ventral diencephalon (DC) (l+r).
For each shape, we took the first 49 non-zero eigenvalues. These were normalized to unit volume, divided by their positional index for balancing noise, and then subjected to PCA. The first 5 PCs of each shape were kept for further analyses. We investigated age and sex influences on PC1 by comparing means in different age classes using Welch's test. A univariate GWAS was performed on the residual of each PC after regression on age, age^2, sex, Euler number, total brain volume, surface area, and the first 10 genetic PCs.
Results:
After PCA on the 49-dimensional LBS of each brain structure, Kaiser's criterion suggested to retain between two and five PCs depending on the structure. PC1 explained already 72%±8% (mean±SD) of the variance, making it a valid univariate proxy of shape.
PC2 explained 6%±3% of the variance leading to 79%±6% as cumulative explained variance. For right and left caudate, at least 91% of the variance was explained by the first two PCs.
GWASs on PC1 replicated in total 77% of all signals from the multivariate LBS GWASs with p < 0.05 after FDR correction in each brain structure. Combining GWAS results on the first two PCs, 95.3% of all signals were replicated at the same significance threshold.
All brain structures, except the 4th ventricle showed significant differences in PC1 with increasing age and all brain structures besides the right cerebellum white matter and right cerebellum cortex showed significant differences in PC1 between male and female individuals. P-values have been corrected for the number of structures and comparisons.
Conclusions:
PC1 accounts for 3/4 of the variance in the Laplace-Beltrami eigenvalues of brain structures. We observed strong age and sex influences on PC1 which makes it an easy and interesting variable to consider in future analyses. As GWAS on PC1 replicated most of the genetic signals from our previous multivariate study, it is a good univariate proxy of shape. By that, a wide range of post-GWAS tools are available for further investigation which are not applicable to the multivariate case.
Genetics:
Genetic Association Studies 2
Lifespan Development:
Aging
Modeling and Analysis Methods:
Multivariate Approaches 1
Task-Independent and Resting-State Analysis
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Subcortical Structures
Keywords:
Aging
Computational Neuroscience
Data analysis
Modeling
Multivariate
Statistical Methods
STRUCTURAL MRI
Sub-Cortical
Univariate
1|2Indicates the priority used for review
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.
I accept
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:
I do not want to participate in the reproducibility challenge.
Please indicate below if your study was a "resting state" or "task-activation” study.
Resting state
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Healthy subjects
Was this research conducted in the United States?
No
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.
Yes
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.
Not applicable
Please indicate which methods were used in your research:
Structural MRI
For human MRI, what field strength scanner do you use?
3.0T
Which processing packages did you use for your study?
Free Surfer
Provide references using APA citation style.
Hibar, D. (2015). Common genetic variants influence human subcortical brain structures. Nature, 520, 224–229.
Primus, S.A. (2024). Beyond Volume: Unraveling the Genetics of Human Brain Geometry [Preprint]. medRxiv, https://doi.org/10.1101/2024.06.25.24309376.
Reuter, M. (2006). Laplace–Beltrami spectra as ‘Shape-DNA’ of surfaces and solids. Computer-Aided Design, 38(4), 342-366.
Wachinger C. (2015). BrainPrint: A discriminative characterization of brain morphology. NeuroImage, 109, 232-248.
No