A Critical Re-evaluation of "Spatial dynamics of brain development and neuroinflammation" by Zhang et al., Nature 2025;647:213–227, DOI: 10.1038/s41586-025-09663-y

This repository contains a comprehensive critical analysis of “Spatial dynamics of brain development and neuroinflammation” by Zhang D. et al. (Nature 2025;647:213–227). The purpose of this work is to provide a rigorous, in-depth, evidence-based evaluation of the scientific claims, methodology, analyses, and interpretations presented in the original publication. By systematically examining the conceptual framework, experimental design, data processing, statistical methods, figure construction, supplementary analyses, biological plausibility, and reproducibility, this commentary aims to clarify significant limitations in the study and provide a corrective perspective for researchers who may rely on the original findings.

The critique demonstrates that the core conclusions of Zhang et al.—specifically the existence of a coordinated “developmental–immune spatial axis” in mid-gestational human brain—are not supported by the underlying evidence. The spatial transcriptomics platform used in the study does not possess the resolution required to infer cell-state transitions, cytokine niches, or novel neuroimmune progenitor populations. The data represent static mixed-cell measurements rather than dynamic developmental processes, undermining claims of spatial or pseudo-temporal transitions. The study also suffers from inconsistent quality-control thresholds, inadequate sampling, batch effects, and the absence of orthogonal validation through histology, immunostaining, smFISH, or protein-level assays.

The computational analyses further exhibit statistical flaws, including uncorrected spatial autocorrelation, unstable gene-module definitions, and inappropriate use of pseudo-trajectory inference. These analytical limitations result in visually compelling but biologically unsupported gradients, trajectories, and ligand–receptor predictions. Supplementary figures reveal additional instability across parameter choices and normalization methods, raising concerns about the robustness and reproducibility of the results.

This Zenodo record provides a fully rewritten, polished, academically rigorous commentary integrating conceptual critique, methodological analysis, statistical evaluation, and figure-by-figure dissection. It also includes detailed discussions of biological plausibility, developmental context, and data-integrity concerns. The goal is to offer the scientific community a clear, balanced, and evidence-focused resource that contextualizes the limitations of the original study and prevents misinterpretation or overextension of its claims.

Researchers working in spatial transcriptomics, developmental neurobiology, neuroimmunology, and data-driven brain mapping may find this commentary valuable for understanding best practices, avoiding common methodological pitfalls, and critically evaluating high-impact spatial omics studies.