Technology Category
Computer Vision: Diffusion Models for VisionCognitive Modeling & Cognitive Systems: Neural Spike CodingMachine Learning: Deep Generative Models & Autoencoders
Tera-MIND: Tera-scale mouse brain simulation via spatial mRNA-guided diffusion
🤖 AI Summary
To address the challenges of large-scale data volume, difficulty in preserving native 3D architecture, and weak modeling of molecular interactions in mouse whole-brain spatial transcriptomics, this paper introduces the first tera-voxel–scale (10¹² voxels) 3D spatial mRNA-guided diffusion model. Methodologically, we propose a boundary-aware patch-based diffusion architecture that integrates patch-wise generation with a 3D gene–gene self-attention mechanism to explicitly model spatially constrained, cooperative gene expression within the native 3D coordinate system. Our contributions are threefold: (1) the first whole-brain–scale, high-fidelity, and reproducible 3D generative simulation of spatial transcriptomes; (2) breakthroughs in both voxel-scale capacity (≥10¹²) and spatial resolution, surpassing limitations of conventional approaches; and (3) accurate recapitulation of the spatial distributions and molecular signatures of glutamatergic and dopaminergic systems—establishing a novel paradigm for deciphering the spatial transcriptional basis of brain function.
Technology Category
Application Category
📝 Abstract
Holistic 3D modeling of molecularly defined brain structures is crucial for understanding complex brain functions. Emerging tissue profiling technologies enable the construction of a comprehensive atlas of the mammalian brain with sub-cellular resolution and spatially resolved gene expression data. However, such tera-scale volumetric datasets present significant computational challenges in understanding complex brain functions within their native 3D spatial context. Here, we propose the novel generative approach $ extbf{Tera-MIND}$, which can simulate $ extbf{Tera}$-scale $ extbf{M}$ouse bra$ extbf{IN}s$ in 3D using a patch-based and boundary-aware $ extbf{D}$iffusion model. Taking spatial transcriptomic data as the conditional input, we generate virtual mouse brains with comprehensive cellular morphological detail at teravoxel scale. Through the lens of 3D $gene$-$gene$ self-attention, we identify spatial molecular interactions for key transcriptomic pathways in the murine brain, exemplified by glutamatergic and dopaminergic neuronal systems. Importantly, these $in$-$silico$ biological findings are consistent and reproducible across three tera-scale virtual mouse brains. Therefore, Tera-MIND showcases a promising path toward efficient and generative simulations of whole organ systems for biomedical research. Project website: $href{http://musikisomorphie.github.io/Tera-MIND.html}{https}$
Problem
Research questions and friction points this paper is trying to address.
Simulate tera-scale mouse brain in 3D
Model molecularly defined brain structures
Analyze spatial gene-gene interactions
Innovation
Methods, ideas, or system contributions that make the work stand out.
Tera-scale mouse brain simulation via diffusion model
Spatial mRNA data guides 3D brain modeling
Patch-based generative approach for teravoxel datasets
🔎 Similar Papers
No similar papers found.
