| Abstract: |
| How genetically encoded molecular patterns shape brain-wide neural wiring remains a fundamental question in neuroscience. To address this, we developed SPERRFY (Spatial Positional Encoding for Reconstructing Rules of axonal Fiber connectivitY), a data-driven framework that integrates connectome data with spatial transcriptome data to infer latent positional gradients underlying neural connectivity. Using mouse brain data from the Allen Brain Atlas, SPERRFY analyzes paired gene-expression profiles at the source and target regions of anatomical projections by canonical correlation analysis. This approach identifies multiple gradient pairs that explain major features of the connectome, spanning both global and local organization. Neural connectivity reconstructed from the inferred gradients shows strong predictive performance and substantially outperforms reconstruction based on physical distance alone. Comparison with null models further supports the biological relevance of the extracted structure. By linking the inferred gradients to individual genes, SPERRFY also provides candidate molecular determinants of wiring specificity. These results extend the classical chemoaffinity concept from topographic circuits to whole-brain connectivity and provide a general framework for decoding genetically encoded design principles of brain architecture. |
|