CoChat
Sign Up
PreprintCited by 0MicroRNA in disease regulationSingle-cell and spatial transcriptomicsNeuroinflammation and Neurodegeneration Mechanisms

Human brain microRNAs exhibit cell-type specificity and age dependence

Open access
Dubnov, Serafima, Laski, Lihen, Zchut, Ido, Avidan, Ofek, Bennett, Estelle R., Greenberg, David S., Turjeman, Adi, Nitzan, Mor +2 more·2026·bioRxiv (Cold Spring Harbor Laboratory)
DOI 10.64898/2026.01.04.697535
Open this paper in CoChat — free accountOpen PDF

Abstract

MicroRNAs play key roles in regulating brain processes ranging from neurogenesis to neurological disease and contribute to many cell-type-specific pathways. However, technical limitations of single-cell small RNA sequencing have confined most studies to bulk analyses of whole-tissue specimens, limiting our capacity to resolve the cell-type specificity of brain microRNAs and to study their roles in depth. To generate a comprehensive, cell-type-resolved atlas of microRNAs in the human brain, we isolated neurons, astrocytes, microglia and oligodendrocytes from fresh neurosurgery-derived brain samples and profiled their small RNA repertoires by RNA sequencing. The results revealed pronounced cell-type-associated differences in microRNA profiles, identified novel cell-type-specific microRNA markers and shed new light on the contribution of their originating genomic loci to cell-type specificity. We further characterized the cell-type dependence of microRNA strand preference and isomiR processing, and identified cell-type-dependent microRNA programs associated with brain ageing. Combined with an accompanying statistical tool for cell-type signal enrichment, our atlas provides a publicly available resource for cell-type-resolved microRNA analysis in the human brain.

Verified 2026-11-06 · 23 references not checked yet

AI snapshot

Written from the paper’s full text.

TL;DR

This paper presents a cell-type-resolved atlas of microRNAs and tRNA fragments from live human brain tissue, using FACS-isolated neurons, astrocytes, microglia, and oligodendrocytes followed by small RNA-sequencing. It identifies cell-type-enriched microRNA and tRF markers, links microRNA cell-type specificity to cell-type-associated regulatory regions, and reports neuronal 5′-tRNA halves elevated in Alzheimer’s disease.

Key findings

  • Across 280 expressed microRNAs, 209 were significantly differentially expressed in pairwise cell-type comparisons: 191 neurons vs. microglia, 163 neurons vs. astrocytes, and 96 microglia vs. astrocytes. Unique enriched microRNAs numbered 53 in neurons, 36 in microglia, and 35 in astrocytes; markers with log-fold change >2 numbered 23 neuronal, 13 microglial, and 6 astrocytic.
  • Cell type was the strongest predictor of microRNA levels (mean explained variance 18.3%), followed by patient ID (12%), while batch, pathology type, sex, and age explained 3.0%, 0.7%, 0.5%, and 1.7%, respectively. MicroRNA profiles clustered by cell type with an NMI of 0.8, versus 0.1 for batch.
  • Oligodendrocyte-enriched microRNAs included 20 significantly enriched relative to all other cell types, with two unique markers after filtering: hsa-miR-181b-5p and hsa-miR-33b-5p.
  • Neurons showed the most distinct tRF profiles: 14 enriched tRF families versus 4 in microglia and none in astrocytes; 10 of the 14 neuronal families were 5′ halves. Nuc-Gly-5′-half, Nuc-Leu-5′-half, and Nuc-Lys-5′-half were significantly upregulated in neurons relative to both astrocytes and microglia, and 72% of expressed 5′-tRNA halves were significantly elevated in Alzheimer’s disease postmortem prefrontal cortex samples versus age-matched controls.

Methods

  • Live human brain samples were obtained during neurosurgery for non-infiltrative tumors (39 samples; 17 females, 22 males; ages 37–85). Tissue was processed with the NuNeX protocol, stained for NeuN, GFAP, IBA1, and separately OLIG2, sorted by FACS, validated by qPCR, and subjected to small RNA-sequencing; 66 samples were sequenced (24 NeuN, 24 IBA1, 18 GFAP, 14 OLIG2).
  • MicroRNA differential expression used DESeq2 normalization and the variancePartition/dream linear mixed-effects framework with the model ~0 + Cell.Type + Batch + Sex + Age + (1|ID), including pairwise cell-type contrasts and an oligodendrocyte-versus-all-other-cell-types contrast.
  • tRFs were collapsed into families by genome origin, amino acid, 5′/3′ side, and cleavage type; batch correction used ComBat-seq/pyComBat; pairwise comparisons used Mann-Whitney tests with Benjamini-Hochberg correction. Alzheimer’s disease tRF analysis re-aligned GSE48552 (n=6 per group) and used DESeq2. MicroRNA-QTL and CRE associations used Fisher’s exact tests.

Limitations

  • The paper states that tumor-related processes could influence small RNA profiles; specimens came only from resections of non-invasive, non-infiltrative tumors external to the brain, and the authors report minimal statistical effect of tumor pathology on microRNA and tRF profiles.
  • The paper states that the OLIG2-positive population was mixed, albeit oligodendrocyte-enriched, and that tRF profiles showed a weaker cell-type effect than microRNAs, with a stronger batch effect requiring rigorous correction; oligodendrocyte tRF families had small effect sizes and may reflect the nucleus permeabilization protocol.
  • Evident from the text: the Alzheimer’s disease tRF validation used only n=6 per group, and RNA integrity values ranged from 1.2 to 7.4, which the authors attribute to rRNA content and formaldehyde fixation rather than degradation.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.64898/2026.01.04.697535

Cites

81 of 104 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 46 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    tRFs differed between cell types, with nuclear 5'-tRFs derived from glycine, lysine, and leucine tRNAs significantly elevated in neurons compared to both microglia and astrocytes.

    “The tRFs also differed between cell types, with nuclear 5'-tRFs derived from glycine, lysine, and leucine tRNAs significantly elevated in neurons compared to both microglia and astrocytes.”
    Quote in the paperSubstantial claim
  • Findingown data

    Neuronal tRFs (5'-tRNA halves) were significantly upregulated in Alzheimer's disease postmortem brain samples compared to age-matched controls.

    “These neuronal tRFs were significantly upregulated in Alzheimer's disease postmortem brain samples compared to agematched controls”
    Quote in the paperSubstantial claim
  • Findingown data

    Many cell-type-enriched miRs are associated with quantitative trait loci (miR-QTLs) located in the enhancer regions of the corresponding cell types.

    “We found that many cell-type-enriched miRs are associated with quantitative trait loci (miR-QTLs) located in the enhancer regions of the corresponding cell types”
    Quote in the paperSubstantial claim
  • Findingown data

    Twenty microRNAs are significantly elevated in oligodendrocytes (log2 fold change > 1, adjusted p ≤ 0.05) compared to all other cell types.

    “Heatmap showing z-scores of 20 miRs, significantly elevated in oligodendrocytes (log2 fold change > 1, adjusted p ≤ 0.05), compared to all other cell types.”
    Quote in the paperSubstantial claim
  • Findingreasoning

    After excluding overlaps, 53, 36, and 35 miRs were uniquely enriched in neurons, microglia and astrocytes, respectively.

    “After excluding these overlaps, we identified 53, 36, and 35 miRs uniquely enriched in neurons, microglia and astrocytes, respectively (Figure 2c).”
    Quote in the paperSubstantial claim
  • May combine two claims. 34
  • Classified as probably not a substantial claim. 29
  • The quote only partly states this claim. 2

  • May combine two claims.

    Findingown data

    The procedure enabled direct comparison of miR and tRF profiles across distinct brain cell populations, revealing unique markers of individual cell types and shared transcripts co-expressed across cell types.

    “This procedure enabled direct comparison of miR and tRF profiles across distinct brain cell populations, revealing both unique markers of individual cell types and shared transcripts co-expressed across cell types.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The atlas revealed multiple differences in miR and tRF expression between the four cell types and identified novel cell-type markers.

    “Our atlas revealed multiple differences in miR and tRF expression between the four cell types and consequently identified novel cell-type markers.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    5'-tRNA halves are higher in neurons compared to glia, including neuronal enrichment of 5'-tRNA halves derived from glycine, leucine and lysine tRNAs.

    “We also report higher levels of 5'-tRNA halves in neurons compared to glia, including a particular neuronal enrichment of 5'-tRNA halves derived from glycine, leucine and lysine tRNAs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    miR profiles clustered according to cell type label, with k-means clustering yielding a normalized mutual information (NMI) score of 0.8 between clusters and annotated cell types.

    “the miR profiles clustered according to their cell type label: k-means clustering resulted in sample division that closely matched the annotated cell types (Figure 2a and Supplementary Figure 2b), yielding a normalized mutual information (NMI) score of 0.8 (Methods).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Comparison of tRF family profiles between the oligodendrocyte-enriched population and other cell types identified four significantly enriched families, but the effect sizes were small.

    “Comparison of tRF family profiles between the oligodendrocyte-enriched population and other cell types identified four significantly enriched families; however, the effect sizes were small (Supplementary Table 11; Supplementary Figure 5d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Neurons and astrocytes featured elevated levels of 24 common miRs compared to microglia; astrocytes and microglia featured elevated levels of 56 common miRs compared to neurons; neurons and microglia demonstrated increased levels of only 5 common miRs compared to astrocytes.

    “Neurons and astrocytes featured elevated levels of 24 common miRs, compared to microglia; astrocytes and microglia featured elevated levels of 56 common miRs, compared to neurons, whereas neurons and microglia demonstrated increased levels of only 5 common miRs, compared to astrocytes (Supplementary table 5).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Collapsing raw tRF counts into family sums reduced noise and increased the variance explained by the cell type factor.

    “Collapsing raw tRF counts into family sums reduced noise and increased the variance explained by the cell type factor (Supplementary Figure 5a, b).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Neurons are particularly enriched in nuclear tRF families aligned to the 5' side of their tRNAs of origin, and three distinct tRF families are candidate markers: Nuc-Gly-5'-half, Nuc-Leu-5'-half and Nuc-Lys-5'-half.

    “Neurons are particularly enriched in nuclear tRF families aligned to the 5' side of their tRNAs of origin, and we identified three distinct tRF families as candidate markers: Nuc-Gly-5'-half, Nuc-Leu-5'-half and Nuc-Lys-5'-half.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Nuclear 5'-derived tRNA halves, enriched in neurons in the atlas, were consistently upregulated in AD, with 72% of expressed 5'-tRNA-halves significantly elevated in AD.

    “Remarkably, nuclear 5'-derived tRNA halves, enriched in neurons in our atlas, were consistently upregulated in AD, with 72% of expressed 5'-tRNA-halves significantly elevated in AD (Figure 6h; Supplementary Table 12).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    A murine aging-related miR study comprising 476 miRs contained markers from all major brain cell types, but only microglial enrichment was statistically significant.

    “We also examined a murine aging-related miR study comprising 476 miRs 69 , and although this set contained markers from all major brain cell types, only microglial enrichment was statistically significant”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The association between microglial miRs and miR-QTLs in microglia-specific enhancers, and between oligodendrocyte-enriched miRs and miR-QTLs in oligodendrocyte-specific enhancers, did not reach statistical significance.

    “The association between microglial miRs and miR-QTLs in microglia-specific enhancers, as well as between oligodendrocyteenriched miRs and miR-QTLs in oligodendrocyte-specific enhancers, did not reach statistical significance (Supplementary Figure 4b; Supplementary Table 7).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There were 191 DE miRs between neurons and microglia, 163 between neurons and astrocytes, and 96 between microglia and astrocytes.

    “Specifically, we identified 191 DE miRs between neurons, and microglia,163 between neurons and astrocytes, 96 between microglia and astrocytes (Figure 2b).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Neurons exhibited the most distinct tRF profiles, with 14 enriched tRF families compared to 4 in microglia and none in astrocytes, resulting in 18 cell-type-enriched families.

    “Neurons exhibited the most distinct tRF profiles (Figure 6b), with 14 enriched tRF families (Figure 6c, 6d, 6e) compared to 4 in microglia (Figure 6c) and none in astrocytes (Figure 6d), resulting in 18 cell-type-enriched families (Figure 6e, Supplementary table 10).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    tRF profiles showed a weaker cell type effect than miRs, with an NMI of 0.04 between k-means clusters and cell type labels, compared to 0.8 for miRs.

    “Overall, tRF profiles showed a weaker cell type effect than miRs, with an NMI of 0.04 between k-means clusters and cell type labels, compared to 0.8 for miRs (Supplementary Figure 5c).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    31 neuron-, 20 microglia-, 21 astrocyte-, and 4 oligodendrocyte-enriched miRs were regulated by 152, 113, 108, and 39 QTLs, respectively.

    “we identified 31 neuron-, 20 microglia-, 21 astrocyte-, and 4 oligodendrocyte-enriched miRs regulated by 152, 113, 108, and 39 QTLs, respectively.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There were 82 miRs elevated in neurons, 115 in astrocytes, and 97 in microglia compared with the other two cell types.

    “This yielded 82 miRs elevated in neurons, 115 in astrocytes, and 97 in microglia (Supplementary Table 5).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Variance partitioning revealed cell type as the strongest predictor of miR levels (mean explained variance 18.3%), followed by patient ID (mean explained variance 12%).

    “Variance partitioning revealed cell type as the strongest predictor of miR levels (mean explained variance 18.3%), followed by patient ID (mean explained variance 12%; Supplementary Figure 2a).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The OLIG2-positive population featured increased levels of the neuronal marker SNAP25, the astrocyte markers AQP4 and ATP13A4, and the microglial marker ALOX5AP.

    “However, the OLIG2-positive population also featured increased levels of the neuronal marker SNAP25, the astrocyte markers AQP4 and ATP13A4 and the microglial marker ALOX5AP (Supplementary Figure 3d-f).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    tRF expression profiles displayed a stronger batch effect than miR profiles, requiring more rigorous correction with ComBat-seq.

    “However, unlike miR profiles, tRFs expression profiles displayed a stronger batch effect, requiring more rigorous correction with ComBat-seq 74 (Supplementary Figure 5b; Methods).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [74] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The 63 miRs associated with multiple sclerosis were predominantly enriched in oligodendrocyte-specific miRs, but also showed significant contributions of astrocyte and microglia markers.

    “In contrast, the 63 miRs associated with multiple sclerosis 66 (MS) were predominantly enriched in oligodendrocyte-specific miRs, but also showed significant contributions of astrocyte and microglia markers”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [66] 2025, Brain microRNAs differentially expressed in age-related cerebral pathologies, Neurobiology of Aging source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Brain tRNA expression patterns vary between cell types.

    “Similarly, brain tRNA expression patterns have been shown to vary between cell types 37 , although cell-type-specific regulation of tRFs remains largely unexplored.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [37] 2016, Expression of microRNA-129-2-3p and microRNA-935 in plasma and brain tissue of human refractory epilepsy, Epilepsy Research source

      Not checked in the free run.

    • Rests on an assumption: Cell-type-specific regulation of tRFs remains largely unexplored.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    tRFs have been implicated in neurological and neurodegenerative conditions including Alzheimer's disease, Parkinson's disease, psychological stress, and neuronal ferroptosis.

    “tRFs have likewise been implicated in a growing number of neurological and neurodegenerative conditions 21,22 , including Alzheimer's disease 23,24 (AD), Parkinson's disease 25 (PD), psychological stress 26 , and neuronal ferroptosis 27 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [21] source

      Not checked in the free run.

    • cites [22] source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Most scRNA-seq protocols rely on poly(T) priming to capture polyadenylated mRNAs, whereas most miRs and all tRFs are not polyadenylated.

    “Additionally, most scRNA-seq protocols rely on poly(T) priming to capture polyadenylated mRNAs, whereas most miRs and all tRFs are not polyadenylated 29 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [29] source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    miRs display cell-type-specific expression across multiple tissues, including the brain.

    “That being said, miRs do display cell-type-specific expression across multiple tissues 35,36 , including the brain 32 .”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [35] Scoyni, F. et al. ciRS-7 and miR-7 regulate ischemia-induced neuronal death via glutamatergic signaling. Cell Reports 43, 113862 (2024). source

      Not checked in the free run.

    • cites [36] source

      Not checked in the free run.

    • cites [32] 2023, High-throughput morphometric and transcriptomic profiling uncovers composition of naïve and sensory-deprived cortical cholinergic VIP/CHAT neurons, The EMBO Journal

      Not checked in the free run.

  • May combine two claims.

    Premisenone given

    The cell-type-specific expression of brain miRs and tRFs remains poorly characterized, limiting functional interpretation of their regulatory roles.

    “However, despite extensive evidence for their involvement in brain function, the cell-typespecific expression of brain miRs and tRFs remains poorly characterized, limiting the functional interpretation of their regulatory roles.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: No human brain single-cell or cell-type-resolved miR datasets exist that simultaneously include all major brain cell types.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    miRs are well-established regulators of key brain processes including neurodevelopment, synaptic plasticity, apoptosis, neuroinflammation, neurodegeneration, and epilepsy.

    “miRs are well-established regulators of key brain processes 4,5 , including neurodevelopment [6][7][8][9][10][11] , synaptic plasticity [12][13][14] , apoptosis 15,16 , neuroinflammation 17 , neurodegeneration 18,19 , and epilepsy 5,20 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [4] source

      Not checked in the free run.

    • cites [5] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Premisenone given

    No robust, commercial scRNA-seq technology currently enables the capture of small RNA profiles, so most studies of brain small RNAs are based on bulk RNA-seq.

    “Since no robust, commercial scRNA-seq technology currently enables the capture of small RNA profiles, most studies of brain small RNAs are based on bulk RNA-seq.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    All primers used were human-specific.

    “All primers were humanspecific.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Normalized expression of neuronal markers SNAP25 and SYNPR was measured across sorted populations (neurons, astrocytes, microglia, and oligodendrocytes) by qPCR.

    “Box plots showing normalized expression of neuronal markers SNAP25 and SYNPR across sorted populations (neurons, astrocytes, microglia, and oligodendrocytes), determined by qPCR.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The study was approved by the Rambam Medical Center institutional review board and the national ethical review board, and all patients signed informed consent forms.

    “The study, including acquisition of cortical brain samples, was approved by the Rambam Medical Center institutional review board and by the national ethical review board (IRB protocol RMB-0713-19).All patients signed informed consent forms.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    The atlas offers an opportunity to deconvolve cell-type signals from bulk-derived miR profiles.

    “In this context, our atlas offers an opportunity to deconvolve cell-type signals from bulk-derived miR profiles.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Nuc-Lys-5'-half was recently reported to regulate neuronal ferroptosis in the context of botulinum intoxication.

    “Interestingly, Nuc-Lys-5'-half was recently reported to regulate neuronal ferroptosis in the context of botulinum intoxication 27 .”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [27] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The miR-33 family is associated with lipid metabolism in the brain, and the miR-181b family was implicated in myelin-assisted axon regeneration.

    “The miR-33 family is associated with lipid metabolism in the brain 58 , and the miR-181b family was implicated in myelin-assisted axon regeneration 59 .”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [58] source

      Not checked in the free run.

    • cites [59] 2024, Mapping the microRNA landscape in the older adult brain and its genetic contribution to neuropsychiatric conditions, Nat Aging source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    A potential limitation of the study is the possible influence of tumor-related processes on the small RNA profiles.

    “On the other hand, a potential limitation of our study is the possible influence of tumor-related processes on the small RNA profiles.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    The distinct neuronal tRF signatures could be explained by cell-type-specific tRF production or degradation rates affected by differential expression of corresponding ribonucleases.

    “which could be explained by cell-type-specific tRF production or degradation rates that are affected by differential expression of the corresponding ribonucleases.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    miR-7 regulates neuronal death and glutamatergic signaling; miR-137 is critical for neural development; miR-935 and miR-129-2-3p are elevated in epileptic brains and plasma.

    “miR-7 regulates neuronal death and glutamatergic signaling 40,41 ; miR-137 is critical for neural development 42 ; miR-935 and miR-129-2-3p are both elevated in epileptic brains and plasma 43 .”
    Quote in the paperNot a substantial claim0 of 4 sources checked
    • cites [40] 2021, Intravitreal injection of anti-miRs against miR-142-3p reduces angiogenesis and microglia activation in a mouse model of laser-induced choroidal neovascularization, Aging source

      Not checked in the free run.

    • cites [41] Gupta, N. et al. miR-142-3p Regulates BDNF Expression in Activated Rodent Microglia Through Its Target CAMK2A. Front. Cell. Neurosci. 14, 132 (2020). source

      Not checked in the free run.

    • cites [42] source

      Not checked in the free run.

    • cites [43] Shan, Y. , Hu, J. , Lv, H. , Cui, X. & Di, W . miR-221 Exerts Neuroprotective Effects in Ischemic Stroke by Inhibiting the Proinflammatory Response. Journal of Stroke and Cerebrovascular Diseases 30, 105489 (2021). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    hsa-miR-181a-5p has been implicated in oligodendrocyte maturation and suggested as a potential epigenetic biomarker of multiple sclerosis.

    “previously implicated in oligodendrocyte maturation 56 and suggested as a potential epigenetic biomarker of multiple sclerosis 57”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [56] source

      Not checked in the free run.

    • cites [57] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The hsa-miR-142 family has previously been associated with microglia in murine brain and linked to neuroinflammation in murine models.

    “previously associated with microglia in murine brain 44 and linked to neuroinflammation in murine models 46,47”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [44] Liang, T. , Liu, Y. , Liu, R. , Shen, X. & Xu, M . miR-221 activates Sox11 to reduce brain injury after intracerebral hemorrhage via inhibiting neuroinflammation. Sci Rep 15, 29643 (2025). source
    • cites [46] 2018, miR-32-5p-mediated Dusp5 downregulation contributes to neuropathic pain, Biochemical and Biophysical Research Communications source
    • cites [47] 2020, MicroRNA-32-5p knockout eliminates lipopolysaccharide-induced depressive-like behavior in mice through inhibition of astrocyte overactivity, Brain, Behavior, and Immunity source

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    miR cell-type specificity may be regulated by miR-QTLs and by miR genes residing within cell-type-associated genomic regions such as cell-type-specific enhancers and other cis-regulatory elements.

    “Furthermore, we propose that miR celltype specificity may be regulated by miR-QTLs and by miR genes residing within cell-type-associated genomic regions such as cell-type-specific enhancers and other cis-regulatory elements.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductown data

    qPCR-determined levels of microglial biomarkers ALOX5AP and CD74 were measured in all three cell types, with microglia values set as controls.

    “Box plot as in h for the microglial biomarkers ALOX5AP and CD74, with microglia values set as controls.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    qPCR-determined levels of SNAP25 and SYNPR were measured in all three cell types, with neuronal levels set as controls.

    “Box plot showing qPCR-determined levels of SNAP25 and SYNPR in all three cell types, with neuronal levels set as controls.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    The dream framework was used to perform differential expression analysis while accommodating fixed and random effects such as cell type and patient ID.

    “we used the dream framework 39 .This approach employs linear mixed-effects modeling to perform differential expression (DE) analysis while accommodating both fixed and random effects, such as cell type and patient ID, respectively.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [39] Grafals-Ruiz, N. et al. MicroRNA-92b targets tumor suppressor gene FBXW7 in glioblastoma. Front. Oncol. 13, (2023). source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Differential expression analysis of cell-type-enriched miRs corrected for batch, pathology type, sex, age and patient ID.

    “To identify cell-type-enriched miRs we performed DE analysis in three pairwise comparisons (neurons vs. microglia, neurons vs. astrocytes and microglia vs. astrocytes), correcting for batch, pathology type, sex, age and patient ID”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    The FACS-sorted NeuN-positive, GFAP-positive, and IBA1-positive populations expressed their respective cell-type markers and not markers of other cell types.

    “NeuN-positive nuclei expressed the neuronal markers SNAP25 and SYNPR (Figure 1g), GFAP-positive nuclei expressed the astrocytic markers AQP4 and ATP13A4 (Figure 1h), and IBA1-positive nuclei expressed the microglial markers ALOX5AP and CD74 (Figure 1i).No cell type expressed markers specific to any other cell type.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    hsa-miR-219a and hsa-miR-338 are known to promote myelin repair, and hsa-miR-23 enhances oligodendrocyte differentiation and myelin synthesis.

    “both known to promote myelin repair 54 , and hsa-miR-23, which enhances oligodendrocyte differentiation and myelin synthesis 55”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [54] source
    • cites [55] 2016, MicroRNA Expression Patterns in Human Astrocytes in Relation to Anatomical Location and Age, J Neuropathol Exp Neurol source

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Normalized expression of astrocyte markers AQP4 and ATP13A4 was measured across sorted populations by qPCR.

    “Same as (d) for astrocyte markers AQP4 and ATP13A4.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims. The quote only partly states this claim.

    Study design and conductown data

    qPCR-determined levels of astrocyte biomarkers AQP4 and ATP13A4 were measured in all three cell types, with astrocyte values set as controls.

    “Box plot as in h for the astrocyte biomarkers AQP4 and ATP13A4, with astrocyte values set as controls.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Normalized expression of microglia markers ALOX5AP and CD74 was measured across sorted populations by qPCR.

    “Same as (d) for microglia markers ALOX5AP and CD74.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    qPCR-measured levels of ENPP2 and MOBP were normalized to GADPH and ACTB transcripts, with oligodendrocyte levels set as controls.

    “qPCR-measured levels of ENPP2 and MOBP, normalized to GADPH and ACTB transcripts, with oligodendrocyte levels set as controls.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    hsa-miR-92b-3p has previously been linked to astrocytes in murine brain and shown to regulate glioblastoma development in human brain.

    “has previously been linked to astrocytes in murine brain 44 and shown to regulate glioblastoma development in human brain 45”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [44] Liang, T. , Liu, Y. , Liu, R. , Shen, X. & Xu, M . miR-221 activates Sox11 to reduce brain injury after intracerebral hemorrhage via inhibiting neuroinflammation. Sci Rep 15, 29643 (2025). source

      Not checked in the free run.

    • cites [45] source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Expression levels of miRs in the human brain are influenced by specific genetic variants termed quantitative trait loci (miR-QTLs).

    “Expression levels of miRs in the human brain were recently reported to be influenced by specific genetic variants, termed quantitative trait loci (miR-QTLs) 60 .”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [60] source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The RIN range most likely reflects varying amounts of rRNA and possible cleavage due to formaldehyde fixation, not low RNA quality due to degradation.

    “The RIN range most likely reflects varying amounts of rRNA present in the samples, in addition to possible cleavage due to the formaldehyde fixation, and not low RNA quality due to degradation.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The RIN range observed reflects varying rRNA and fixation, not low RNA quality due to degradation.
52 of 52

  • The paper relies on this without support.

    No human brain single-cell or cell-type-resolved miR datasets exist that simultaneously include all major brain cell types.

    “Currently, to the best of our knowledge, no human brain single-cell or cell-typeresolved miR datasets exist, that simultaneously include all major brain cell types.”

  • The paper relies on this without support.

    Cell-type-specific regulation of tRFs remains largely unexplored.

    “although cell-type-specific regulation of tRFs remains largely unexplored.”

  • The paper relies on this without support.

    The RIN range observed reflects varying rRNA and fixation, not low RNA quality due to degradation.

    “The RIN range most likely reflects varying amounts of rRNA present in the samples, in addition to possible cleavage due to the formaldehyde fixation, and not low RNA quality due to degradation.”
3 of 3

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • DE analysis claims pathology adjustment absent from its model

    Minor

    The results say the pairwise miR comparisons corrected for pathology type, but the stated DE model has no pathology term.

  • Pathology is reported as a variance factor but omitted from the stated model

    Minor

    The results give a mean proportion of miR variance explained by pathology type, but the variance-partitioning model specified in Methods contains no pathology term.

2 of 2

Cited works that were retracted

Works this paper cites that were later retracted, from the retraction notices publishers deposit. Major when the notice came out before this paper was published.

  • Cites a retracted work: 10.3892/mmr.2019.10903

    Info

    The cited work was retracted; the date of the notice is not on record.

    Cited work 10.3892/mmr.2019.10903 · retraction date not known · this paper published 2026-01-04

1 of 1

Cites

1 retracted

Flagged first, then by citations

Research works printed but not linked: 4

  1. 10.1101/2024.10.31.24316513 · DOI · Paper page
  2. 10.1101/2024.10.31.24316513 · DOI · Paper page
  3. 10.1101/2024.05.22.595292 · DOI
  4. Chen, J. et al. Bioinformatic analyses and validated experiments reveal an aging hallmark gene set and protective miR of coronary artery disease. Sci Rep 15, 32102 (2025). · DOI

Cited by

Still being checked.

Related

Authors
Dubnov, S., Laski, L., Zchut, I., Avidan, O., Bennett, E. R., Greenberg, D. S., Turjeman, A., Nitzan, M., Paldor, I., & Soreq, H.
Affiliations
  • Hebrew University of Jerusalem (IL)
  • The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, 9190401, Israel
  • Shaare Zedek Medical Center (IL)
Corresponding author
Soreq, Hermona
ORCID
Venue
bioRxiv (Cold Spring Harbor Laboratory)
Publisher
openRxiv
Published
2026-01-04
DOI
10.64898/2026.01.04.697535
OpenAlex
W7118182759
Semantic Scholar
1edccfc86e5b8b59061635885d73245de9ee4461
Cited by
0
References
81
Type
posted content
Language
en
Open access
Licence
Locations
bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc)
Funders
  • Teva Pharmaceutical Industries
  • Hebrew University of Jerusalem
  • Israel Science Foundation (835/23)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › MicroRNA in disease regulation
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
Keywords
microRNA, human brain, cell type specificity, brain aging, neurons, astrocytes, microglia, oligodendrocytes, isomiRs, microRNA processing, cell type markers, small RNA profiling

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-06.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-06.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review