#32 · Single-cell
Mesoscale fingerprints and constrained dynamics characterize functional brain networks
The dominant thread is graph-based characterization of brain organization at the network level. A signed multilayer community-detection framework produces mesoscale functional connectome fingerprints, evaluated on 810 HCP subjects across sessions. Separately, diffusion-prior Laplacian smoothing of dynamic connectivity reveals reduced functional flexibility in cocaine use disorder, with structural constraints built into the dynamics model. The published item is less directly network-based: gsModel uses spatial transcriptomics cell-type composition to classify macaque sulci versus gyri at over 90% accuracy. What is new across these is the explicit incorporation of sign, multilayer structure, and structural priors into functional network inference.
Published papers
Cross-species single-cell atlas of the striatum defines cell-type and subregion disease vulnerabilities.
Human-mouse snRNA-seq striatum atlas linking subregional cell types to opioid, antipsychotic and Huntington's disease vulnerabilities.
The cellular and molecular basis of gyrus-sulcus specialization in the primate brain.
Introduces gsModel, a deep learning framework using spatial transcriptomics cell-type composition to classify macaque sulci vs gyri (>90% accuracy), linked to…
Preprints
Structurally Constrained Brain Network Dynamics Reveal Reduced Functional Flexibility in Cocaine Use Disorder
Diffusion-prior Laplacian smoothing of dynamic connectivity reveals reduced network flexibility in cocaine use disorder.
Topic history
| Date | Rank |
|---|---|
| #32 |