A Study of Neuroimaging and Genetic Non-linear Associations For Brain Development
Poster No:
686
Submission Type:
Abstract Submission
Authors:
Jingyu Liu1, Bhaskar Ray2, Dawn Jensen3, Jiayu Chen1, Vince Calhoun4
Institutions:
1GSU, Atlanta, GA, 2Georgia State. University, Atlanta, GA, 3Georgia State University, Austell, GA, 4GSU/GATech/Emory, Atlanta, GA
First Author:
Co-Author(s):
Introduction:
During the transition from childhood to adulthood, the human brain undergoes significant developmental changes that underlie cognitive maturation, driven by a complex interplay of environmental and genetic factors. Despite abundant studies on neuroimaging for brain development, the specific genetic influences remain unclear. Using brain age estimation, we categorized individuals (N = 7,435, aged 9–10 years) from the Adolescent Brain and Cognitive Development (ABCD) cohort into groups exhibiting either accelerated or delayed brain maturation. This study examines brain imaging and genetic associations within this subpopulation to identify the genetic influences on brain development.
Methods:
To uncover the genetic associations with neuroimaging features underlying brain development, we integrated genetic information (1,030 SNPs associated with IQ from a large GWAS study) with brain features extracted from structural and resting-state functional MRI images. These brain features include 100 gray matter networks derived from independent component analyses, representing regional gray matter density, 152 brain morphological measures from FreeSurfer, and 160 PCA components of resting state functional network connectivity matrix]. Sparse canonical correlation analysis (SCCA) was implemented as the base model to capture linear imaging-genetic relationships, and a deep neural network model (DNN) was designed to capture complex non-linear connections between genetics and brain characteristics. As in Figure 1 the DNN model architecture combines auto-encoders and multilayer perceptrons to reduce dimensionality and refine latent genetic and imaging representations to maximize their correlation. We tested three loss functions for DNN models: a) direct Latent Representation distance (JLR), b) JLR and group classification (JLR-GC), and c) group supervised contrastive learning and JLR (SC-JLR). Lost functions of B and C leveraged the information of brain development groups (Accelerated group vs. Delayed group). To prevent overfitting, dropout, batch normalization, L2 regularization, and cross-validation were applied, ensuring robust and generalizable results.
Results:
SCCA failed to identify significant linear relationships between genetic and imaging data, yielding poor performance on validation (r = 0.04), and test datasets (r = 0.02). In contrast, the non-linear DNN models extract significant imaging genetic correlations with r=0.07, 0.09 and 0.12 in validation dataset for the JLR, JLR_GC and SC-JLR models, respectively. The correlations are 0.06,0.09,0.11 in the test dataset for the three models. SC-JLR extracted the most significant latent correlations, revealing decreased gray matter density in the middle temporal and inferior frontal regions and increased connectivity in the inferior frontal and hippocampus areas, prominent in the accelerated group (see Figure 2). Genetic features highlighted SNPs in CADM2, CALN1, NREP, TNRC6A, and MARK3, with predominant expression in the frontal cortex, cerebellar hemisphere. The brain latent representation showed significant group differences, while genetic latent representation did not have group differences.
Conclusions:
In this research, we integrated multimodal brain imaging and genetic data to explore associations during brain development. The SCCA baseline failed to identify significant relationships due to its limitation to linear modeling, while all three non-linear models captured robust associations. The SC-JLRM model, utilizing supervised contrastive learning extracted the most significant correlated latent representations between genetics and brain characteristics. The identified correlated latent representations of genetic and brain imaging data enhance our understanding of the role of genetic influence on brain maturation and lay a foundation for further exploration of genetic influences on brain development.
Genetics:
Genetic Association Studies 1
Genetic Modeling and Analysis Methods
Lifespan Development:
Early life, Adolescence, Aging 2
Modeling and Analysis Methods:
Multivariate Approaches
Other Methods
Keywords:
Development
Modeling
Multivariate
Other - Genetic effect
1|2Indicates the priority used for review
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?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
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. van Drunen, L.,et al., (2024). Genetic and environmental influences on structural brain development from childhood to adolescence: A longitudinal twin study on cortical thickness, surface area, and subcortical volume. Developmental cognitive neuroscience, 68, 101407.
3. Cole, J. H., et al., (2018). Brain age predicts mortality. Molecular psychiatry, 23(5), 1385-1392.
4. Ray, B., et al (2021). Multimodal Brain Age Prediction with Feature Selection and Comparison. The 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) 3858-3864.
5. Ray, B., et al. (2024). Replication and Refinement of Brain Age Model for adolescent development. In 2024 IEEE International Symposium on Biomedical Imaging (ISBI) 1-5.
6. Ray, B.,et al., (2024). Adolescent brain maturation associated with environmental factors: a multivariate analysis. Frontiers in Neuroimaging, 3, 1390409.
7. Savage, J.E., et al (2018). Genome-wide association meta-analysis in 269,867 individuals identifies new genetic and functional links to intelligence. Nature Genetics 50, 912–919
8 Xu, L., et al (2009). Source‐based morphometry: The use of independent component analysis to identify gray matter differences with application to schizophrenia. Human brain mapping, 30(3), 711-724.
9. Du Y., et al. NeuroMark: An automated and adaptive ICA based pipeline to identify reproducible fMRI markers of brain disorders. NeuroImage: Clinical. 2020;28