Neuroimaging-derived Brain Biological Age Correlates with Multi-organ Proteome-based Biological Age
Poster No:
887
Submission Type:
Abstract Submission
Authors:
Junhao Wen1, Christos Davatzikos2, Jingyue Wang1, Mehrshad Saadatinia1, Filippos Filippos Anagnostakis1, Sarah Ko1
Institutions:
1Columbia University, New York, NY, 2Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA
First Author:
Co-Author(s):
Introduction:
Brain morphological changes are shaped by both normal aging and pathological processes. Magnetic resonance imaging (MRI)-based brain age prediction provides individualized metrics, such as brain biological age gap (BAG), to quantify these changes and assess deviations from the normative aging trajectory (Bashyam et al., 2020). However, brain aging is a complex biological process interconnected with multiple organ systems and spans various temporal and spatial scales (Wen et al., 2024). This study investigated the relationship between the neuroimaging-derived brain BAG (PhenoBAG) and 11 organ-specific proteome-based BAGs (ProtBAG) using multimodal MRI and plasma proteomics data from the UK Biobank.
Methods:
The brain PhenoBAG was extracted from our previous study using gray matter volumetric measures. For the 11 multi-organ ProtBAGs, we predicted the chronological age using three machine learning models: linear SVR, lasso regression, and neural network (NN) and organ-specific over-expressed proteins. To fairly compare their performance, we curated a training set of 4589 people without any pathologies (CN) for the training/validation/test dataset. We randomly sampled 500 CN for the independent test dataset. The rest of the patients with any ICD-10 diagnosis were set as independent data and were only applied to the trained model after model selections by the nested CV. We reported the mean absolute error (MAE) and Pearson's correlation (r) between the predicted age and the chronological age for the CN-independent test dataset.
Results:
When fitting the organ-specific proteins to three AI/ML models [i.e., Lasso regression, support vector regressor (SVR), and neural network (NN)], we observed marginal variability in model performance, with no single model consistently outperforming the others (4.33<MAE<10.19; 0.09<r<0.69). For instance, the Lasso model outperformed NN and SVR for the hepatic ProtBAG (P-value < 2.27x10-6, though the standard t-test may be permissive in a complex cross-validation setting). In addition, the brain PhenoBAG (MAE=4.47) achieved comparable model performance with the brain ProtBAG (MAE=4.86; two-sample t-test P-value=0.088) (Fig. 1). We then compute the genetic correlation (Bulik-Sullivan et al. 2015) and phenotypic correlation between the brain PhenoBAG and the 11 ProtBAGs. We observed significant phenotypic correlations (PC) between the brain PhenoBAG and the brain (PC=0.08), eye (PC=-0.05), heart (PC=0.06), and skin (PC=0.05) ProtBAGs. However, while not statistically significant, genetic correlations did not consistently align with the phenotypic correlations between the two BAGs (e.g., the brain PhenoBAG vs. eye ProtBAG) (Fig. 2).
·Fig 2
·Fig 1
Conclusions:
The current study generates 11 multi-organ proteome-based aging clocks and links them to neuroimaging-derived brain BAG, highlighting the cross-organ interconnectedness between the brain and body organ systems.
Genetics:
Genetic Association Studies 2
Lifespan Development:
Aging 1
Keywords:
Aging
MRI
Statistical Methods
Other - Multi-organ
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.
Wen, J., Tian, Y.E., Skampardoni, I., Yang, Z., Cui, Y., Anagnostakis, F., Mamourian, E., Zhao, B., Toga, A.W., Zalesky, A. and Davatzikos, C., 2024. The genetic architecture of biological age in nine human organ systems. Nature Aging, pp.1-18.
Bulik-Sullivan, B.K., Loh, P.R., Finucane, H.K., Ripke, S., Yang, J., Schizophrenia Working Group of the Psychiatric Genomics Consortium, Patterson, N., Daly, M.J., Price, A.L. and Neale, B.M., 2015. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nature Genetics, 47(3), pp.291-295.