The study, published in BMC Neuroscience by a team at Xi’an Jiaotong University led by corresponding author Shiquan Sun, generated single-nucleus RNA sequencing data from adult mouse brains under deliberately varied experimental designs. Rather than testing one variable at a time in isolation, the researchers dissected the individual and interactive effects of three major workflow choices: whether tissue was harvested post-mortem or in vivo, whether nuclei were fixed in methanol or left fresh, and whether the library was sequenced on the widely used 10x Genomics platform or the SeekOne digital droplet system. Each of these decisions is made routinely in labs around the world, often with little consideration of how they might reshape the final dataset.

The most striking finding concerns the standard post-mortem harvest, the default approach in most single-nucleus studies because it is technically simpler than extracting nuclei from living tissue. The researchers found that post-mortem collection triggers a rapid and selective loss of metabolically active glutamatergic neuron subpopulations, the excitatory neurons that form the backbone of cortical computation. This loss is not a uniform shrinkage of all cell types; it preferentially eliminates specific vulnerable subpopulations, meaning the resulting cell composition no longer faithfully reflects the living brain. Intriguingly, the global transcriptome appeared largely preserved, which means the damage can hide beneath standard quality-control metrics that look only at overall gene detection.

That selective vulnerability carries a troubling implication for brain atlases. The study reports that while the global transcriptomic profile survives the post-mortem interval, transcripts related to advanced cognitive functions become obscured. In other words, a researcher comparing healthy and diseased brains might attribute differences in cognition-related gene expression to disease, when in fact the differences stem from how long the tissue sat before nuclei were extracted. Because glutamatergic neurons are central to nearly every model of cortical function, from memory circuits to the excitatory-inhibitory balance implicated in autism and schizophrenia, their disproportionate sensitivity to harvest stress is a bias that no atlas builder can afford to ignore.

The second variable the team examined was fixation. Methanol fixation is popular because it stabilizes samples for shipping and storage, allowing labs to process tissue collected at distant sites and to pause experiments mid-workflow. The benchmarking showed that methanol fixation minimally altered cell-type compositions, which is reassuring for anyone worried that the chemical treatment might preferentially destroy certain nuclei. However, the treatment was not transcriptionally neutral. It introduced cell-type-specific gene expression shifts, preferentially stabilizing structural and core functional transcripts in both neurons and glia. The practical consequence is that the measured abundance of particular genes depends on whether the sample was fixed, even when the underlying biology is identical.

Perhaps the most consequential discovery, however, involves the sequencing platform itself, a variable most researchers assume is merely a matter of cost and throughput. The team found that platform choice shaped biological interpretation in a profound way: the 10x Genomics and SeekOne digital droplet platforms captured largely non-overlapping differential expression signatures in response to harvest stress. When the same biological perturbation was profiled on the two systems, each platform reported a different set of genes as changing. The 10x platform showed a bias toward detecting synaptic transcript degradation, while SeekOne leaned toward preserved metabolic and glial programs. Two labs studying the same question with the same tissue could therefore reach substantively different molecular conclusions based on platform alone.

This platform dependence strikes at the heart of the reproducibility crisis in single-cell genomics. Large-scale initiatives, including efforts to build comprehensive cell atlases of the human and mouse brains, aggregate data generated on different instruments and with different preservation protocols. If the differential expression signatures triggered by a stressor are largely platform-specific, then integrating datasets without accounting for these effects risks conflating technical artifacts with genuine biological signal. The study’s authors argue that platform choice must be treated as a critical, non-neutral variable rather than an interchangeable commodity, and that meta-analyses should model platform effects explicitly rather than assuming they wash out in large datasets.

To separate artifact from biology, the researchers deployed a battery of quantitative benchmarks. Their evaluation toolkit included clustering concordance measures such as the Adjusted Rand Index and Normalized Mutual Information, cell-type separation metrics like the cell-type average silhouette width, and integration diagnostics including batch silhouette widths and Local Inverse Simpson’s Index variants that quantify how well cell types mix or remain distinct after computational integration. By applying these metrics across the factorial experimental designs, the team could attribute specific distortions to specific workflow steps. This framework, they emphasize, is not just a catalog of problems but a practical recipe: it gives other labs a way to audit their own pipelines and to design studies in which technical variables are measured and controlled rather than left unexamined.

The study also maps the cellular geography of vulnerability in fine detail. Drawing on established brain cell taxonomy, the team characterized glutamatergic neuron subclasses including intratelencephalic and extratelencephalic projecting cortical neurons, layer 6b near-projecting corticothalamic neurons, dentate gyrus immature glutamatergic neurons, and glutamatergic populations of the hypothalamus, thalamus, and midbrain, alongside GABAergic neurons, oligodendrocyte lineage cells, and astrocyte-ependymal cells. This granularity matters because the selective loss of metabolically active glutamatergic subpopulations is not visible at the level of broad cell classes. An analysis that lumps all excitatory neurons together would miss the fact that particular subclasses are disappearing, and with them the transcripts that define their specialized functions.

For the field at large, the message is a call for radical transparency in methods reporting. The authors conclude that precise reporting and control of workflow variables are prerequisites for accurate biological discovery in single-cell transcriptomics, particularly as the community moves toward integrative brain cell atlases meant to serve as durable community resources. Journals and consortia increasingly demand detailed metadata, and this study supplies a concrete reason why: harvest method, fixation chemistry, and sequencing platform each leave measurable fingerprints on the data, and those fingerprints can masquerade as biology. Until such variables are standardized or explicitly modeled, the researchers warn, differences between datasets should be interpreted with caution, and claims of disease-related or experience-related transcriptional changes in the brain should be validated across harvest conditions and platforms before being accepted as fact.

Subject of Research: Technical biases in single-nucleus RNA sequencing of the mouse brain

Article Title: A systematic multi-factorial benchmarking reveals critical technical biases in single-nucleus RNA sequencing of the brain

Article References: A systematic multi-factorial benchmarking reveals critical technical biases in single-nucleus RNA sequencing of the brain. (n.d.). https://doi.org/10.1186/s12868-026-01058-7

Image Credits: AI Generated

DOI: 10.1186/s12868-026-01058-7

Keywords: single-nucleus RNA-seq, benchmarking, brain cell atlas, glutamatergic neurons, post-mortem harvest, methanol fixation, sequencing platform bias, 10x Genomics, SeekOne, transcriptomics, neuroscience, reproducibility

The scale of the dataset matters because statistical power in single-cell genetics is notoriously hard to come by. Most variants identified by genome-wide association studies sit in non-coding regions of the genome, where they are thought to act as regulatory elements that dial gene expression up or down. Bulk tissue studies, which average signals across millions of mixed cells, have linked many of these variants to specific genes, but they cannot resolve which of the brain’s dozens of cell types actually carry out the regulation. By aggregating nuclei from each donor into cell-type-specific pseudobulk profiles and fitting linear mixed models that account for population structure, the researchers detected significant eQTLs for 14,258 genes, with 981 genes showing regulatory effects specific to a cell class and 857 specific to a subclass.

The number of detectable eGenes varied enormously across cell types, and the authors are candid about why. Excitatory neurons yielded 10,913 eGenes at the class level, while endothelial cells yielded only 414. At the subclass level, layer 2/3 intratelencephalic excitatory neurons produced 8,812 eGenes compared with just 1,683 in layer 6b excitatory neurons. Much of this variation tracks cell type abundance and average sequencing depth per cell, a reminder that rare but biologically important populations remain underpowered even in a study of this magnitude. Concordance between cell-type-level and bulk-level effect sizes also rose with cell abundance, with neurons showing markedly higher agreement than non-neuronal cells.

To validate the biology behind the statistics, the team examined whether cell-type-specific open chromatin regions, identified from single-cell ATAC-seq data, were enriched around the lead eQTL variants of the matching cell types. They were, with particularly strong enrichment for glial cells, consistent with the idea that regulatory variants act through cell-type-specific chromatin landscapes. Fine-mapping with a probabilistic model narrowed candidate causal variants to credible sets of roughly ten to thirty variants per locus. An independent check against a massively parallel reporter assay in stem-cell-derived excitatory neurons showed that fine-mapped regulatory variants in that cell type had the strongest association with measured allelic effects, lending functional weight to the catalog.

The most striking findings emerge when the regulatory atlas is crossed with disease genetics. Using stratified linkage disequilibrium score regression, the researchers showed that regulatory variants from neurons are enriched for the heritability of neuropsychiatric traits, with schizophrenia showing the broadest signal, followed by bipolar disorder and major depressive disorder. But astrocytes and oligodendrocytes were also enriched for these psychiatric traits, and oligodendrocyte precursor cells carried signals for schizophrenia and depression. Neurodegenerative traits told a different story: Alzheimer’s and Parkinson’s disease heritability was enriched in microglia but not in neuronal subclasses, echoing the established role of the brain’s resident immune cells in neurodegeneration.

Colocalization analysis, which tests whether the same variant drives both gene regulation and disease risk, pinpointed specific genes and cell types. The analysis identified 46 genes colocalizing with Alzheimer’s disease risk, 22 with major depressive disorder and 46 with schizophrenia. Up to 18 of these genes were detected exclusively in the single-nucleus data and had been invisible to previous bulk RNA-seq studies. For schizophrenia, many colocalized genes appeared only in excitatory or inhibitory neurons, including FUT9, ACE and FURIN, while DRD2 and FAM171A1 were specific to excitatory neurons and ERBB4 and SP4 to inhibitory neurons. For Alzheimer’s, beyond 16 genes in immune cells driven largely by microglia, the team found nine genes in oligodendrocytes, 12 in astrocytes and six in neurons, with microglial genes enriched for amyloid-beta formation and oligodendrocyte genes for complement activation, hinting at an immune-regulatory role for these myelin-producing cells.

Resolution proved decisive. The gene CNTN4, involved in cell adhesion, colocalized with schizophrenia risk only in layer 6 corticothalamic excitatory neurons, a signal that vanished when neuronal subclasses were collapsed into broad classes. Similarly, SORL1, a well-known Alzheimer’s gene, colocalized with disease risk in microglia but not in perivascular macrophages, distinguishing two cell populations that bulk analyses would have merged. The team also tackled the thorny statistical problem of proving that a regulatory effect is genuinely cell-type-specific rather than merely undetected elsewhere. Using a multivariate Bayesian meta-analysis with a novel composite hypothesis test, they directly estimated the posterior probability that an effect is non-zero in one cell type and zero in all others. A standout example is EGFR, which carries at least two independent regulatory programs, one in astrocytes and one in oligodendrocytes; only the astrocytic signal, led by variant rs74504435 with a specificity posterior of 0.946, colocalizes with Alzheimer’s risk.

Because the donor cohort spans the entire postnatal lifespan, from birth to over a century old, the study could also ask whether genetic regulation itself changes over development. Using a supervised pseudotime trajectory anchored to donor age, the researchers tested whether variant effects on expression shift along the trajectory, fitting negative binomial mixed models at the single-nucleus level. They identified 2,073 genes with dynamic eQTLs, ranging from 1,364 in excitatory neurons to just nine in oligodendrocyte precursor cells, and these dynamic genes were enriched for developmental processes such as neuron generation and cell junction organization. Crucially, dynamic eGenes in excitatory neurons, inhibitory neurons and oligodendrocytes were enriched for genes with disease colocalization signals, and a handful, including CLU and BIN1 for Alzheimer’s and FAM171A1 for schizophrenia, showed dynamic regulation and disease colocalization in the same cell class.

Finally, the atlas reaches beyond local regulation. Trans-eQTL analysis, testing variants more than five megabases from their target genes, uncovered 1,655 genes under distal regulation, with limited overlap across cell classes and replication rates of 66 to 95 percent in an independent cohort. The team identified four trans-regulatory hubs centered on the transcriptional regulators SUPT3H, RUNX2 and ZNF160, the largest of which controls nine downstream targets in oligodendrocytes. In astrocytes, mediation analysis supported a mechanistic chain in which a variant regulates the developmental gene RERE locally and the neurodevelopmental gene AUTS2 in trans, with both signals weakly colocalizing with schizophrenia risk. Together, these layers of cis, trans and time-varying regulation sketch a dynamic, cell-by-cell picture of how the genome builds and maintains the human brain, and they offer a growing list of mechanistic targets for disorders that have long resisted explanation. The authors emphasize that expanding sample sizes and capturing rarer cell types will be the next frontier, alongside integrating multi-omic data and diverse ancestries to move brain genetics toward precision medicine.

Subject of Research: Cell-type-specific genetic regulation of gene expression in the human prefrontal cortex and its role in neuropsychiatric and neurodegenerative disease risk

Article Title: Single-nucleus atlas of cell-type specific genetic regulation in the human brain

Article References: Zeng, B., Yang, H., N. M, P., Venkatesh, S., Mathur, D., Auluck, P., Bennett, D. A., Marenco, S., Haroutunian, V., PsychAD Consortium, Hong, A., Li, A. Z., He, C., Gupta, C., Dillard, C., Porras, C., Casey, C., McClung, C. A., Spencer, C., … Roussos, P. (2026). Single-nucleus atlas of cell-type specific genetic regulation in the human brain. Nature Genetics, 58(10), 2525-2534. https://doi.org/10.1038/s41588-026-02733-5

Image Credits: AI Generated

DOI: 10.1038/s41588-026-02733-5

Keywords: single-nucleus RNA-seq, eQTL, human brain, Alzheimer's disease, schizophrenia, microglia, astrocytes, oligodendrocytes, gene regulation, GWAS, colocalization, neurodevelopment

The research focused on a well-established mouse model of drug-resistant temporal lobe epilepsy, in which kainic acid, a potent excitotoxin, is microinjected into the amygdala. This model is prized by epilepsy researchers because it recapitulates several defining features of the human condition, including unilateral hippocampal damage, spontaneous recurrent seizures, and a stubborn resistance to conventional anti-seizure medications. Two weeks after inducing status epilepticus, the prolonged seizure state that marks the transition from injury to chronic epilepsy, the researchers dissected the ipsilateral hippocampus, the side of the brain exposed to the initial insult, and prepared it for single-nucleus RNA sequencing.

Single-nucleus RNA-seq is a technique that reads out the gene expression profile of individual cell nuclei, allowing scientists to catalogue which genes are active in each of the thousands of distinct cell types that make up a tissue. Rather than averaging signals across a whole chunk of brain, which would bury cell-specific changes in noise, the method resolves the molecular fingerprint of every neuron, astrocyte, oligodendrocyte, microglial cell and endothelial cell separately. In this study, the team processed samples from epileptic and control mice on the 10X Genomics platform, ultimately sequencing 34,737 nuclei that together represented all the major cellular subtypes of the hippocampus. The resulting dataset constitutes what the authors describe as an atlas of transcriptomic change for this model, a reference map that other researchers can mine for years to come.

The first major finding was that gene dysregulation in epilepsy is not a uniform phenomenon but is sharply partitioned by both cell subtype and anatomical subfield. The hippocampus is organised into anatomically distinct regions, including the dentate gyrus and the CA1, CA2 and CA3 fields, each with its own circuitry and vulnerability. Neurons across the dentate gyrus, CA1 and CA3 all showed increases in activity-regulated genes, the immediate-early programmes that neurons deploy in response to strong stimulation. But beneath that shared signature, the subfields diverged dramatically. Dentate gyrus neurons displayed the greatest degree of differential gene expression of any neuronal population, and their altered genes were enriched for processes connected to synaptogenesis, synaptic plasticity and NMDA receptor signalling, the very molecular machinery that governs how connections between neurons are formed, strengthened and pruned.

That enrichment is significant because the dentate gyrus is widely regarded as a gatekeeper of the hippocampal circuit, filtering the flow of information from the entorhinal cortex into the rest of the hippocampus. In temporal lobe epilepsy, this gate is thought to fail, allowing hyperexcitable activity to propagate unchecked. The new data suggest that the failure is accompanied, and perhaps driven, by an extensive molecular programme of synaptic remodelling within dentate granule cells themselves. If the cells are actively rebuilding their synapses and recalibrating their NMDA receptor signalling two weeks after the initial insult, then the epileptic hippocampus is not passively deteriorating but is engaged in a maladaptive reconstruction, one that may lock in the hyperexcitable state rather than resolve it.

The CA3 and CA1 fields told different stories altogether. Transcriptional profiles from CA3 glutamatergic neurons pointed towards mechanical and injury responses, consistent with the particular vulnerability of this subfield to excitotoxic damage in seizure disorders. CA1 neurons, by contrast, carried signatures enriched for neuroinflammation and wound healing, suggesting that this region responds to the epileptic state less like injured wiring and more like tissue mounting a repair-and-defence programme. The implication is striking: within a structure measuring only a few millimetres, the same disease process produces at least three distinct cellular narratives, each defined by geography. Any therapy designed to treat the epileptic hippocampus as a single homogeneous target may therefore be addressing only one part of a much more heterogeneous problem.

Glial cells, the non-neuronal supporting cells of the brain, proved equally instructive. Among all glial populations, astrocytes showed the most extensive differential gene expression in epileptic mice. Their profiles revealed a mixed reactive state, simultaneously exhibiting pro-inflammatory and anti-inflammatory responses alongside altered metabolic signatures. Astrocytes are far more than passive scaffolding; they regulate neurotransmitter clearance, control local blood flow, supply metabolic substrates to neurons and help define the extracellular environment in which synaptic transmission occurs. The finding that epileptic astrocytes occupy a conflicted, dual-polarity reactive state hints at a cell type caught between containment and contribution, potentially both buffering the seizure disorder and facilitating it. Disentangling those two roles is now an explicit target for future work.

Perhaps the most forward-looking component of the study was its cell-to-cell communication analysis, which used the gene expression data to predict how signalling conversations between cell types change in epilepsy. The analysis predicted extensive remodelling of hippocampal signalling across cell subtypes, including alterations in glutamatergic transmission and, notably, changes in latrophilin-mediated ligand-receptor pairing between dentate gyrus and CA2 neurons. Latrophilins are adhesion G-protein-coupled receptors that help specify which neurons form synapses with which partners during development and plasticity. Evidence that this pairing system is reorganised in the epileptic hippocampus suggests that the disease may be physically rewiring the connectome, changing not just the strength of existing synapses but the very pattern of who talks to whom. That kind of structural reorganisation could help explain why drug-resistant epilepsy is so stable and so difficult to reverse once established.

The clinical stakes of this work are considerable. Roughly a third of people with epilepsy do not respond adequately to available anti-seizure medications, and temporal lobe epilepsy is among the most pharmacologically stubborn forms of the disorder. Current drugs largely suppress symptoms by damping neuronal excitability without addressing the underlying network adaptations that sustain the disease. By providing a cell-type-resolved catalogue of those adaptations, the new atlas offers a systematic way to identify targets for genuine disease modification, molecules that could interrupt the pathological rewiring rather than merely quiet the electrical storm. It also supplies potential biomarkers, gene signatures detectable in specific cell populations that could be used to track disease progression or treatment response in preclinical trials.

The study, led by Javier Villegas-Salmerón with Niamh M. C. Connolly and David C. Henshall as joint senior authors, was funded by Research Ireland through the Centre for Research Training in Genomics Data Science and the FutureNeuro Research Ireland Centre for Translational Brain Science, with additional support from the European Regional Development Fund and Science Foundation Ireland. The work was conducted under the appropriate EU and Irish animal research approvals. Its authors are careful to note that the data come from a mouse model at a single chronic time point, and that translating findings from sequenced nuclei to human patients will require further validation. Even so, the message of the atlas is hard to ignore: epilepsy is a disease of molecular geography, in which every neighbourhood of the hippocampus tells its own story of injury, inflammation and reconstruction. Reading those stories cell by cell may be the surest route yet towards therapies that do more than mute the symptoms of a brain that has rewired itself against its own interests.

Subject of Research: Single-nucleus transcriptomic profiling of hippocampal cell types in a mouse model of drug-resistant temporal lobe epilepsy

Article Title: Single-nucleus transcriptomic analysis reveals subfield-specific cell-to-cell synaptic reorganisation of the mouse hippocampus in focal temporal lobe epilepsy

Article References: Single-nucleus transcriptomic analysis reveals subfield-specific cell-to-cell synaptic reorganisation of the mouse hippocampus in focal temporal lobe epilepsy. (n.d.). https://doi.org/10.1186/s12967-026-08978-2

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08978-2

Keywords: temporal lobe epilepsy, single-nucleus RNA-seq, hippocampus, kainic acid model, dentate gyrus, astrocytes, synaptic remodelling, latrophilin signalling, neuroinflammation, gene expression, CA1, CA3