Related Experiment Video
Updated: Jun 13, 2026
05:45
Generation and Downstream Analysis of Single-Cell and Single-Nuclei Transcriptomes in Brain Organoids
Published on: March 29, 2024
Cross-scale spatially-aware generative modeling of transcriptomic programs underlying neurodegenerative brain
1Department of Computer Engineering, Karaikal Polytechnic College, Karaikal, Puducherry, India.
Arxiv
|June 12, 2026
Summary
This study introduces a new generative model to link gene expression in the brain to neurodegenerative disease patterns. The framework accurately predicts regional brain vulnerability in Alzheimer's disease, offering insights into disease mechanisms.
Area of Science:
- Computational Neuroscience
- Genomics
- Neurobiology
Background:
- Neurodegenerative diseases like Alzheimer’s disease show distinct regional brain vulnerability patterns.
- Current imaging-transcriptomic studies often use correlational methods, limiting understanding of molecular mechanisms driving neurodegeneration.
- The spatial selectivity of neurodegeneration remains poorly understood.
Purpose of the Study:
- To develop a cross-scale, spatially-aware generative framework to model transcriptomic programs underlying cortical neurodegeneration.
- To link regional gene expression organization with patterns of neurodegeneration.
- To investigate the biological mechanisms behind spatially selective neurodegeneration.
Main Methods:
- Utilized the Allen Human Brain Atlas for regional transcriptomic profiles (910 genes, 68 cortical regions).
- Constructed neurodegenerative vulnerability maps from ADNI FreeSurfer cortical thickness data (comparing cognitively normal and Alzheimer's disease subjects).
- Employed a variational generative architecture with graph-based spatial regularization to model gene expression-neurodegeneration links.
Main Results:
- The framework achieved high prediction accuracy for regional neurodegenerative vulnerability (explained variance = 0.8604).
- Demonstrated significant spatial correlation between predicted and observed cortical degeneration (r = 0.9439, p < 0.001).
- Identified structured latent representations linking transcriptomic organization to disease susceptibility.
Conclusions:
- Biologically constrained generative modeling can effectively bridge molecular organization and macroscale neurodegeneration.
- The framework provides a foundation for spatially-aware generative neurobiology and computational neuroscience.
- This approach enhances understanding of the molecular basis of spatially selective neurodegeneration.
