← Back to 单细胞与多组学分析

Single-cell multiomic human brain atlas reveals regulatory drivers of cortical regionality

A strong atlas paper showing how transcriptome, chromatin accessibility, and spatial context can be joined to infer region-specific regulatory programs.

Nat CommunsnRNA-seqsnATAC-seqspatialGRN

Full Citation

Palmer CR, Song J, Yang B, Chen CJ, Diep D, Conklin K, et al. Single-cell multiomic human brain atlas reveals regulatory drivers of cortical regionality. Nat Commun. 2026;17:3051.

Study typeOpen-access human atlas study integrating single-nucleus RNA-seq, single-nucleus ATAC-seq, and spatial transcriptomics across cortical regions.
IdentifierNo PMID listed
DOI10.1038/s41467-026-69368-2

Background and Question

Single-cell RNA-seq identifies cell states, but disease interpretation often requires regulatory context. Multiomic atlases connect gene expression to chromatin accessibility and spatial architecture, making them a better template for mechanism discovery.

Research question

Which cell-type compositions and gene-regulatory networks distinguish major human cortical regions, and how can combined RNA, ATAC, and spatial data reveal drivers of cortical regionalization?

Methods and Evidence Chain

Sampling

Nine distinct human cortical regions from six postmortem donors were anatomically dissected and profiled.

Modalities

Combined single-nucleus RNA-seq, single-nucleus ATAC-seq, and spatial transcriptomics.

Analysis

Built cell taxonomies, regional proportion maps, differential expression, chromatin-accessibility profiles, and gene-regulatory network inferences.

Validation logic

Connected molecular programs with cortical anatomy and spatial organization.

1

Sampling

Nine distinct human cortical regions from six postmortem donors were anatomically dissected and profiled.

2

Modalities

Combined single-nucleus RNA-seq, single-nucleus ATAC-seq, and spatial transcriptomics.

3

Analysis

Built cell taxonomies, regional proportion maps, differential expression, chromatin-accessibility profiles, and gene-regulatory network inferences.

4

Validation logic

Connected molecular programs with cortical anatomy and spatial organization.

Key Results

Atlas scale

The study profiled cortical regions spanning frontal, motor, sensory, temporal, auditory, angular, and visual cortex.

Cell composition

Cell classes and subclasses showed region-specific abundance and expression patterns.

Regulatory programs

Integrated ATAC and RNA data nominated regulatory drivers of cortical regional identity.

Spatial anchoring

Spatial transcriptomics helped ground single-nucleus findings in tissue context.

Mechanism Interpretation

The work models regional identity as a hierarchy: anatomical region shapes cell-subclass composition; chromatin accessibility constrains which genes can be expressed; transcription-factor networks execute region-specific programs; spatial context preserves laminar and neighborhood information.

Mechanism / workflow schematic

Mermaid source is included so the website can render the diagram in supported browsers.

flowchart LR
  A[Anatomical sampling] --> B[snRNA-seq]
  A --> C[snATAC-seq]
  A --> D[Spatial transcriptomics]
  B --> E[Cell taxonomy]
  C --> F[Regulatory elements]
  D --> G[Tissue context]
  E --> H[Regional identity model]
  F --> H
  G --> H
  H --> I[Testable regulatory targets]

Clinical and Translational Relevance

Clinical relevance

Although this is a neuroscience atlas rather than a skin study, the methodological logic transfers directly to scar and wound research: combine scRNA, scATAC, and spatial transcriptomics to separate cell abundance shifts from true regulatory rewiring.

Translational value

For a scar/wound lab, the paper is a blueprint for moving from descriptive cluster labels to targetable regulatory networks and spatially localized repair niches.

Limitations and Critique

Tissue state

Postmortem brain tissue differs from live disease biopsies and may contain donor and preservation effects.

Causality

Regulatory-network inference generates hypotheses rather than direct proof.

Transferability

Brain regionalization mechanisms are not directly skin mechanisms.

Cost

Multiomic atlas designs require careful sample planning and budget control.

Reviewer-style critique

The paper is a model of atlas design because it integrates modalities instead of treating single-cell RNA as sufficient. Its weakness is the usual atlas limitation: it is excellent at nominating regulatory programs but needs perturbation systems to prove which nodes control phenotype.

Practical Next Research Actions

Action 1

Design a keloid/wound atlas with paired lesional, edge, adjacent normal, and time-course healing biopsies.

Action 2

Use scATAC to distinguish fibroblast activation caused by cell abundance from stable chromatin remodeling.

Action 3

Add spatial transcriptomics to map invasive keloid edges, vascular niches, nerves, and inflammatory zones.

Action 4

Prioritize transcription factors and ligand-receptor pairs for perturbation in primary skin organoid or fibroblast models.

Evidence-quality judgment

High atlas-quality evidence for human cortex biology; high methodological relevance for single-cell multi-omics workflows.