CoChat
Sign Up
Cited by 1Alzheimer's disease research and treatmentsDementia and Cognitive Impairment ResearchNeuroinflammation and Neurodegeneration Mechanisms

Magnetic resonance microscopy maps widespread effects of Alzheimer’s disease on brain structures and behavior in mice

Open access
Tian, Yuqi, Hornburg, Kathryn, Austin, Wyatt, Cook, James J., Qi, Yi, White, Leonard E., Killmar, John T., Kaczorowski, Catherine C. +3 more·2026·Nature Neuroscience
DOI 10.1038/s41593-025-02199-4PMC PMC13185701
Open this paper in CoChat — free account

Abstract

Alzheimer's disease has widespread effects on brain structure, function and behavior, but we lack a systematic dissection of its impact across hundreds of forebrain and brainstem regions. Here, using diffusion tensor MRI at 25 µm, we mapped the global consequences of mutations in APP and PSEN1 across 231 regions of interest (ROIs) in male and female 5×FAD BXD hybrid mice at 14 months. Over half of the ROIs change in volume along rostrocaudal and mediolateral axes of the CNS, with unexpected swelling in the neocortex, hippocampus and amygdala of up to 10%, counterbalanced by shrinkage in the thalamus, brainstem and most white matter tracts. Yet, total brain volume is unaltered. Variation in individual ROI volumes is highest in females. Differences in fear acquisition and contextual memory performance covary with volumes of several regions and can have opposite polarities between cases and controls. These structural benchmarks establish a foundation for testing therapeutic interventions in preclinical trials.

AI snapshot

Written from the paper’s full text.

TL;DR

This study used 25 μm diffusion tensor MRI to map 5×FAD Alzheimer’s disease mutations across 231 CNS regions in genetically diverse AD-BXD F1 mice at 14 months. It found widespread regional volume changes—rostral enlargement and caudal shrinkage—without altering total brain size, with effects modulated by genetic background and sex and correlated with fear learning/memory measures.

Key findings

  • Of 231 ROIs, 137 had significant volume changes as a percentage of total brain volume at FDR <0.05; total brain volume did not differ significantly between cases and controls (475.7 ± 3.7 mm³ vs 477.8 ± 6.0 mm³, s.e.).
  • Effects followed a rostrocaudal gradient: telencephalic grey matter structures such as neocortex, hippocampus, and amygdala increased in volume by up to 10%, while thalamus, brainstem, and most fiber tracts decreased; the corticospinal tract lost up to 8.1% volume and 11% fractional anisotropy.
  • Genetic background explained 23% of variance in fractional volume across all ROIs and 20% for the 137 significant ROIs; sex explained about 7%, with 123 ROIs significant in females and 93 in males.
  • Behavioral covariation was detected: 29 ROIs correlated with contextual fear acquisition at PBH <0.05 and 9 with contextual fear memory; case-control correlation polarity differences were nominally significant for 46 ROIs with fear acquisition but none survived FDR correction.

Methods

  • Case-control design using four AD-BXD F1 hybrid types (AD-BXD32, AD-BXD65, AD-BXD77, AD-BXD101) at 14 months: 57 mice total, including 34 5×FAD cases and 23 controls, with 26 males and 31 females.
  • Ex vivo diffusion tensor MRI at 25 μm isotropic resolution with 61 diffusion angles, gadolinium contrast, and automated alignment to CCFv3/DMBA atlases; 230 ROIs plus whole brain were delineated bilaterally, with right-left coefficient of variation averaging 1.8%.
  • Behavioral testing included contextual fear acquisition and contextual fear memory for 52 of 57 mice; a subset of AD-BXD77 F1 mice underwent light-sheet microscopy and stereology for Aβ and NeuN. Analyses used ANOVA on fractional ROI volumes with Benjamini-Hochberg or Bonferroni correction, Levene’s and Welch’s tests, and Spearman/Pearson correlations.

Limitations

  • The paper states that case-control correlation analyses had modest sample sizes (31 cases and 21 controls) and were not well powered; no case-control correlation differences survived FDR correction, and no transgene-by-background or sex-by-transgene interactions survived multiple-test correction.
  • Stated technical caveats include possible antibody penetration/specificity issues in light-sheet stereology, an implausible increase in NeuN+ cells in visual cortex that may reflect sampling error or nonspecific labeling, and possible common antibody penetration patterns driving Aβ–NeuN correlations.
  • The text notes that 5×FAD does not model late-stage human AD well, particularly tau tangle formation, and that translational relevance of the model remains controversial; the study is also cross-sectional at 14 months and uses only four AD-BXD F1 types for the main volumetric analysis.

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

Integrity signals

Registry

Crossref · 10.1038/s41593-025-02199-4

Cites

103 of 103 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

56 of 74 cited sources checked · 16 supported · 0 flagged · accuracy not yet measured

  • Study design and conductown data

    The right-left dissection replicability provides sufficient precision to detect much smaller volume differences than in previous work.

    “This level of right-left dissection replicability provides sufficient precision to detect much smaller volume differences than in previous work33.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [33] Crombé, A., Nicolas, R., Richard, N., Tourdias, T. & Hiba, B. High B-value diffusion tensor imaging for early detection of hippocampal microstructural alteration in a mouse model of multiple sclerosis. Sci. Rep. 12, 12008 (2022). source

      The Materials and methods (Data analysis; page unavailable) describes delineating hippocampal regions on MRI images, not right-left physical dissection. The Results (Quality control—SNR and CNR; page unavailable) discusses image quality, but our copy provides no right-left dissection replicability measure or comparison of detectable volume differences with previous work. Searches for those terms found no supporting passage. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
  • Study design and conductcitation

    Light-sheet stereology gained accuracy because the diffeomorphic alignment used corrects for tissue distortions during whole-brain tissue clearing.

    “Light-sheet stereology also gained accuracy because the diffeomorphic alignment that we use corrects for tissue distortions that occur during whole-brain tissue clearing31.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [31] Mansour, H. et al. The Duke Mouse Brain Atlas: MRI and light sheet microscopy stereotaxic atlas of the mouse brain. Sci. Adv. 11, eadq8089 (2025). source
  • Interpretationown data

    This set highlights many sensory, perceptual, and motor systems critical for fear conditioning.

    “Collectively, this set highlights many sensory, perceptual, and motor systems critical for fear conditioning54-57.”
    Quote in the paperNot a substantial claim2 of 4 sources checked
    • cites [54] Gusel’nikova, V. V. & Korzhevskiy, D. E. NeuN as a neuronal nuclear antigen and neuron differentiation marker. Acta Naturae 7, 42–47 (2015). source

      The Introduction discusses NeuN as a neuronal marker and mentions some motor neurons, but it does not describe a set of sensory, perceptual, and motor systems or their role in fear conditioning. Searches for fear conditioning and for sensory and motor systems found no passage addressing that claim. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [55] Greco, J. A. & Liberzon, I. Neuroimaging of fear-associated learning. Neuropsychopharmacology 41, 320–334 (2016). source
    • cites [56] Antoniadis, E. A. & McDonald, R. J. Amygdala, hippocampus and discriminative fear conditioning to context. Behav. Brain Res. 108, 1–19 (2000). source
    • cites [57] Keifer, O. P. Jr., Hurt, R. C., Ressler, K. J. & Marvar, P. J. The physiology of fear: reconceptualizing the role of the central amygdala in fear learning. Physiology 30, 389–401 (2015). source
  • Findingreasoning

    Our positive findings contrast sharply with negative findings of the MODEL-AD consortium and other recent studies that have also used T1/T2 MRI and DTI methods.

    “Our positive findings contrast sharply with negative findings of the MODEL-AD consortium16 and other recent studies that have also used T1/T2 MRI and DTI methods7,44 (Supplementary Tables 2 and 3).”
    Quote in the paperSubstantial claim3 of 3 sources checked
    • cites [16] Oblak, A. L. et al. Comprehensive evaluation of the 5×FAD mouse model for preclinical testing applications: a MODEL-AD study. Front. Aging Neurosci. 13, 713726 (2021). source

      The Materials and Methods section (Magnetic Resonance Imaging; page unavailable) confirms that this MODEL-AD study acquired T2-weighted MRI images. The cited passages do not report negative T1/T2 MRI or DTI findings: they describe T2 imaging for the PET/MRI work, while the Results report PET-detected changes. Thus the citation supports the study identification and T2 use, but not the attributed negative findings or use of T1 and DTI. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [7] Murdy, T. J. et al. Leveraging genetic diversity in mice to inform individual differences in brain microstructure and memory. Front. Behav. Neurosci. 16, 1033975 (2022). source
    • cites [44] Andersen, J. V. et al. Hippocampal disruptions of synaptic and astrocyte metabolism are primary events of early amyloid pathology in the 5×FAD mouse model of Alzheimer’s disease. Cell Death Dis. 12, 954 (2021). source

      The Materials and methods sections describe metabolic, proteomic, and electron-microscopy analyses, not T1/T2 MRI or DTI. The Discussion passage on regional brain metabolism does not report negative MRI findings or discuss the MODEL-AD consortium. Searches for those terms found no passage in our copy that addresses the attributed claim; page numbers were not available. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
  • Findingcitation

    Regions with higher mean activity and glucose demands tend to match those ROIs that swell.

    “Regions with higher mean activity and glucose demands tend to match those ROIs that swell.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [49] Evangelio, M., García-Amado, M. & Clascá, F. Thalamocortical projection neuron and interneuron numbers in the visual thalamic nuclei of the adult C57BL/6 mouse. Front. Neuroanat. 12, 27 (2018). source

      The Results section on dLGN and LP complex volume reports anatomical volume differences, not swelling of regions with higher mean activity or glucose demand. The Discussion section on volume speculates that the larger left dLGN might reflect lateralized sensory use, but does not test the claimed relationship. Searches for glucose metabolism, activity and swelling found no passage addressing it; page numbers were not available in our copy. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [50] Eimer, W. A. & Vassar, R. Neuron loss in the 5×FAD mouse model of Alzheimer’s disease correlates with intraneuronal Aβ42 accumulation and caspase-3 activation. Mol. Neurodegener. 8, 2 (2013). source

      The Results sections on Caspase-3 activation and Aβ42 deposition describe where neuronal degeneration occurs in 5XFAD mice, not which regions swell. Searches for regional activity, glucose demand, and swelling found no passage in our copy linking these measures. (LLM, conf=0.75)

      Experimental — accuracy not yet measured
  • Classified as probably not a substantial claim. 30
  • May combine two claims. 57

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

    Interpretationcitation

    As in human families with PSEN1 mutations, the effects may be restricted to the prodromal phase and be followed by serious volume reduction and neuron death.

    “As in human families with PSEN1 mutations, this may be restricted to the prodromal phase and be followed by serious volume reduction and neuron death57.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [57] Keifer, O. P. Jr., Hurt, R. C., Ressler, K. J. & Marvar, P. J. The physiology of fear: reconceptualizing the role of the central amygdala in fear learning. Physiology 30, 389–401 (2015). source

      The available passages from the paper’s Abstract discuss fear learning, amygdala circuits, and genetic variants associated with fear and PTSD. Searches for PSEN1 mutations, a prodromal phase, volume reduction, and neuronal death found no passage addressing the attributed claim. (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Most studies have used only the C57BL/6J strain, which is known to be relatively resistant to the transgene effects.

    “One source of differences may be that most studies have used only the C57BL/6J strain10, known to be relatively resistant to the transgenes effects.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [10] Neuner, S. M., Heuer, S. E., Huentelman, M. J., O’Connell, K. M. S. & Kaczorowski, C. C. Harnessing genetic complexity to enhance translatability of Alzheimer’s disease mouse models: a path toward precision medicine. Neuron 101, 399–411 (2019). source

      The Discussion (‘Improving rigor and reproducibility in preclinical studies’; page unavailable) says traditional studies use a single inbred strain, but does not establish that most use C57BL/6J specifically. The Discussion (‘The role of modifier genes in normal aging and resilience to AD’; page unavailable) says the B6 background appears to attenuate the cognitive effects of 5XFAD mutations, supporting the resistance point for those effects rather than transgene effects generally. (LLM, conf=0.83)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Mutations in both APP and PSEN1 cause aberrant APP cleavage and clearance that lead to neuroinflammation, activation of astrocytes and microglia, and changes in bioenergetics.

    “Mutations in both APP and PSEN1 cause aberrant APP cleavage and clearance4,47 that lead to neuroinflammation, the activation of astrocytes and microglia, and changes in bioenergetics44,48.”
    Quote in the paperNot a substantial claim3 of 4 sources checked
    • cites [4] Ringman, J. M. et al. Genetic heterogeneity in Alzheimer disease and implications for treatment strategies. Curr. Neurol. Neurosci. Rep. 14, 499 (2014). source

      The 'Amyloid Precursor Protein' and 'The Presenilins' sections (pages not supplied) describe how APP and PSEN1 mutations alter APP cleavage and Aβ production. The 'Autosomal dominant forms of AD' section distinguishes reduced Aβ clearance in late-onset AD from increased Aβ synthesis in PSEN1 mutation carriers; it does not establish aberrant clearance for both mutations. The later inflammation, glial-activation, and bioenergetics claims carry separate citations. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [47] Macdonald, I. R. et al. Early detection of cerebral glucose uptake changes in the 5×FAD mouse. Curr. Alzheimer Res. 11, 450–460 (2014). source

      The Introduction (page not provided) describes familial-AD mutation mouse models with activated microglia, neuroinflammation, mitochondrial abnormalities, and possible changes in glucose metabolism. It does not mention astrocyte activation, and it presents the metabolic link as a possibility rather than an established consequence. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [44] Andersen, J. V. et al. Hippocampal disruptions of synaptic and astrocyte metabolism are primary events of early amyloid pathology in the 5×FAD mouse model of Alzheimer’s disease. Cell Death Dis. 12, 954 (2021). source

      The Introduction (page unavailable) links Aβ accumulation to astrocyte changes and early metabolic changes in Alzheimer’s disease. The Discussion (page unavailable) reports altered astrocyte metabolism in 5xFAD mice, but does not establish the claimed chain from aberrant APP cleavage and clearance to neuroinflammation and microglial activation. The cleavage-and-clearance portion also has separate citations in the citing sentence. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [48] Shin, L. M. & Liberzon, I. The neurocircuitry of fear, stress, and anxiety disorders. Neuropsychopharmacology 35, 169–191 (2010). source

  • May combine two claims.

    Interpretationcitation

    The use of sophisticated multiangle methods enhances contrast and precision of delineations, explaining positive results.

    “2. the use of sophisticated multiangle methods that enhance contrast and precision of delineations31;”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [31] Mansour, H. et al. The Duke Mouse Brain Atlas: MRI and light sheet microscopy stereotaxic atlas of the mouse brain. Sci. Adv. 11, eadq8089 (2025). source

      The Results sections on diffusion tensor data and the average MRH atlas report that more angular samples help resolve crossing fibers and that multiple MRI contrasts improve recognition of cytoarchitectural boundaries. Figure 2 also attributes improved image contrast and clearer borders to precision registration and averaging. The paper does not attribute both improved contrast and more precise delineation to a single multiangle method. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S1 (asserted), S3 (load-bearing).

    Interpretationcitation

    Polarities of effects are linked to differences in metabolic rates.

    “One plausible idea is that polarities of effects are linked to differences in metabolic rates49-51.”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [49] Evangelio, M., García-Amado, M. & Clascá, F. Thalamocortical projection neuron and interneuron numbers in the visual thalamic nuclei of the adult C57BL/6 mouse. Front. Neuroanat. 12, 27 (2018). source

      The Introduction discusses differences in neuron numbers between mouse strains, and the Discussion describes differences in labeling among visual thalamic nuclei; neither links the polarity of effects to metabolic rates. Searches for the quoted phrase and for metabolic rates or metabolism found no passage in our copy addressing that proposed link. No page numbers were provided for these passages. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [50] Eimer, W. A. & Vassar, R. Neuron loss in the 5×FAD mouse model of Alzheimer’s disease correlates with intraneuronal Aβ42 accumulation and caspase-3 activation. Mol. Neurodegener. 8, 2 (2013). source

      The Results and Discussion passages address neuron loss, Aβ42 accumulation, and Caspase-3 activation in 5XFAD mice, not polarities of effects or metabolic rates. Searches for the quoted phrase and for metabolic-rate differences found no passage addressing the proposed link; page numbers were not available in our copy. (LLM, conf=0.75)

      Experimental — accuracy not yet measured
    • cites [51] Busser, J., Geldmacher, D. S. & Herrup, K. Ectopic cell cycle proteins predict the sites of neuronal cell death in Alzheimer’s disease brain. J. Neurosci. 18, 2801–2807 (1998). source

      The Discussion compares how quickly neurons die during development and in Alzheimer’s disease, but does not link that difference to metabolic rates. Searches for metabolic rates and related terms found no passage addressing the proposed explanation for polarities of effects. The indexed passages provide no page numbers. (LLM, conf=0.73)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Our findings align with the work of Gurdon and colleagues who detected decreased NeuN label in the hippocampus.

    “Our findings align with the work of Gurdon and colleagues8 who detected decreased NeuN label in the hippocampus.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [8] Gurdon, B. et al. Detecting the effect of genetic diversity on brain composition in an Alzheimer’s disease mouse model. Commun. Biol. 7, 605 (2024). source

  • May combine two claims.

    Premisecitation

    These ROIs have been previously associated with fear memory and align with work at the circuit level.

    “These ROIs have been previously associated with fear memory56,59-66 and align with work at the circuit level62.”
    Quote in the paperNot a substantial claim7 of 9 sources checked
    • cites [56] Antoniadis, E. A. & McDonald, R. J. Amygdala, hippocampus and discriminative fear conditioning to context. Behav. Brain Res. 108, 1–19 (2000). source
    • cites [59] Venkataraman, A. & Dias, B. G. Expanding the canon: an inclusive neurobiology of thalamic and subthalamic fear circuits. Neuropharmacology 226, 109380 (2023). source
    • cites [60] Frontera, J. L. et al. The cerebellum regulates fear extinction through thalamo–prefrontal cortex interactions in male mice. Nat. Commun. 14, 1508 (2023). source
    • cites [61] Penzo, M. A. et al. The paraventricular thalamus controls a central amygdala fear circuit. Nature 519, 455–459 (2015). source
    • cites [62] Ratigan, H. C., Krishnan, S., Smith, S. & Sheffield, M. E. J. A thalamic–hippocampal CA1 signal for contextual fear memory suppression, extinction, and discrimination. Nat. Commun. 14, 6758 (2023). source
    • cites [63] Sun, Y., Gooch, H. & Sah, P. Fear conditioning and the basolateral amygdala. F1000Res https://doi.org/10.12688/f1000research.21201.1 (2020). source
    • cites [64] Bernier, B. E. et al. Dentate gyrus contributes to retrieval as well as encoding: evidence from context fear conditioning, recall, and extinction. J. Neurosci. 37, 6359–6371 (2017). source
    • cites [65] Zhang, Y., Wang, Z., Ju, J., Liao, J. & Zhou, Q. Elevated activity in the dorsal dentate gyrus reduces expression of fear memory after fear extinction training. J. Psychiatry Neurosci. 46, E390–E401 (2021). source
    • cites [66] Hernández-Rabaza, V. et al. The hippocampal dentate gyrus is essential for generating contextual memories of fear and drug-induced reward. Neurobiol. Learn. Mem. 90, 553–559 (2008). source

  • May combine two claims.

    Premisecitation

    Cortex and hippocampus have higher glucose utilization than globus pallidus, as do the three exceptional nuclei in thalamus, hypothalamus, and midbrain that all increase in volume-the LGD, SCH, and ICC.

    “For example, cortex and hippocampus have higher glucose utilization than globus pallidus, as do the three exceptional nuclei in thalamus, hypothalamus, and midbrain that all increase in volume—the LGD, SCH, and ICC (Fig. 4a,b)49,50.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [49] Evangelio, M., García-Amado, M. & Clascá, F. Thalamocortical projection neuron and interneuron numbers in the visual thalamic nuclei of the adult C57BL/6 mouse. Front. Neuroanat. 12, 27 (2018). source

      The Results section on dLGN and LP volume reports measurements and a left–right difference in dLGN volume, not an increase in volume across the three named nuclei. Searches for glucose utilization, globus pallidus, SCH, and ICC found no passage addressing the claimed comparisons. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [50] Eimer, W. A. & Vassar, R. Neuron loss in the 5×FAD mouse model of Alzheimer’s disease correlates with intraneuronal Aβ42 accumulation and caspase-3 activation. Mol. Neurodegener. 8, 2 (2013). source

      The Results section discusses neuron loss and Caspase-3 activation in the 5XFAD cortex and subiculum, not regional glucose utilization or volume increases. Searches for glucose utilization, globus pallidus, and the LGD, SCH, and ICC found no passage addressing the claimed comparisons. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Both LGD and LPT are crucial components of the visual system and have dense reciprocal connections with visual cortex, but have radically different physiology and functions, and only the LGD has high numbers of GABAergic interneurons.

    “Both nuclei are crucial components of the visual system and have dense reciprocal connections with visual cortex, but these nuclei have radically different physiology and functions, and only the LGD has high numbers of GABAergic interneurons52.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [52] Park, C. et al. Stress granules contain RBFOX2 with cell cycle-related mRNAs. Sci. Rep. 7, 11211 (2017). source

      The Introduction and Discussion passages concern Rbfox proteins and stress granules, not the visual nuclei described in the claim. Searches for LGD, LPT, reciprocal visual-cortex connections, and GABAergic interneurons found no passage addressing those points. No page numbers were available in our copy. (LLM, conf=0.78)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Previous work has reported covariation of the loss of cortical layer 5 neurons in 5×FAD mice with levels of Aβ42 deposition.

    “Previous work has reported covariation of the loss of cortical layer 5 neurons in 5×FAD mice with levels of Aβ42 deposition53.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [53] Kim, K. K., Adelstein, R. S. & Kawamoto, S. Identification of neuronal nuclei (NeuN) as Fox-3, a new member of the Fox-1 gene family of splicing factors. J. Biol. Chem. 284, 31052–31061 (2009). source

      The NeuN/Fox-3 identification section describes neuronal staining, and the cross-reactivity section examines anti-NeuN staining; neither addresses neuron loss in 5×FAD mice. Searches for cortical layer 5 neuron loss, 5×FAD, and Aβ42 deposition found no passage reporting the claimed covariation. Page numbers are unavailable in our copy. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    A pipeline built on ANTs registered the 66 volumes into a common space to correct for small displacements caused by eddy currents in the magnet.

    “Once scans were completed, a pipeline built on ANTs90 registered the 66 volumes into a common space to correct for small displacements caused by eddy currents in the magnet, which vary between volumes.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [90] St-Jean, S., Coupé, P. & Descoteaux, M. Non Local Spatial and Angular Matching: enabling higher spatial resolution diffusion MRI datasets through adaptive denoising. Med. Image Anal. 32, 115–130 (2016). source

      Section 3.4 (p. 10) says ANTs was used to register white-matter masks to in-vivo datasets, not to register 66 scan volumes into a common space. Section 5.7 (p. 24) discusses eddy-current correction as a preprocessing step but does not say the ANTs registration performed it. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

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

    Study design and conductcitation

    The contextual fear conditioning protocol used matches that previously used in the AD-BXD F1s.

    “The protocol used matches that previously used in the AD-BXD F1s10 83.”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [10] Neuner, S. M., Heuer, S. E., Huentelman, M. J., O’Connell, K. M. S. & Kaczorowski, C. C. Harnessing genetic complexity to enhance translatability of Alzheimer’s disease mouse models: a path toward precision medicine. Neuron 101, 399–411 (2019). source
    • cites [83] Sasani, T. A. et al. A natural mutator allele shapes mutation spectrum variation in mice. Nature 605, 497–502 (2022). source

      The Methods section on construction of BXD lines describes breeding, and the Methods section on whole-genome sequencing describes tissue collection and genetic analysis. Neither describes contextual fear conditioning or compares its protocol with one used in AD-BXD F1s (searched: “contextual fear conditioning AD-BXD F1s” and “fear conditioning protocol behavioral testing Alzheimer BXD F1”). (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    Delineations of the dorsal lateral geniculate nucleus (LGD) of AD-BXD F1s have mean unilateral volumes of approximately 0.35 ± 0.04 SD mm3, whereas previous estimates of adult B6D2F1 hybrid from celloidin-sectioned Nissl-stained material are 0.32 ± 0.03 mm3 with correction for shrinkage.

    “For example, delineations of the dorsal lateral geniculate nucleus (LGD) of AD-BXD F1s have mean unilateral volumes of approximately 0.35 ± 0.04 SD mm3, whereas our previous estimates of adult B6D2F1 hybrid from celloidin-sectioned Nissl-stained material are 0.32 ± 0.03 mm3 with correction for shrinkage94.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [94] Seecharan, D. J., Kulkarni, A. L., Lu, L., Rosen, G. D. & Williams, R. W. Genetic control of interconnected neuronal populations in the mouse primary visual system. J. Neurosci. 23, 11178–11188 (2003). source

      Split into 2 atoms — 0 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 1 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    Eleven ROIs have nominally significant correlations with both behaviors, including key sensory regions (VCP, VCS, and MGD) linked to early symptoms of AD in humans.

    “Eleven ROIs have nominally significant correlations with both behaviors (Supplementary Table 10), including key sensory regions (VCP, VCS, and MGD) linked to early symptoms of AD in humans71-74.”
    Quote in the paperNot a substantial claim3 of 4 sources checked
    • cites [71] Hong, S., Baek, S. H., Lai, M. K. P., Arumugam, T. V. & Jo, D. G. Aging-associated sensory decline and Alzheimer’s disease. Mol. Neurodegener. 19, 93 (2024). source

      The Background describes visual and auditory processing deficits at early stages of AD, and the Visual impairment section links visual-cortex pathology to AD-related visual disruption. The Auditory system section identifies the medial geniculate nucleus as an auditory pathway region, but our copy does not establish that VCP, VCS, and MGD individually are linked to early AD symptoms in humans. The eleven ROI correlations are attributed to the citing paper’s Supplementary Table 10, not to this source. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [72] Murphy, C. Olfactory and other sensory impairments in Alzheimer disease. Nat. Rev. Neurol. 15, 11–24 (2019). source
    • cites [73] Zhang, N. K., Zhang, S. K., Zhang, L. I., Tao, H. W. & Zhang, G. W. Sensory processing deficits and related cortical pathological changes in Alzheimer’s disease. Front. Aging Neurosci. 15, 1213379 (2023). source

      The Introduction (page unavailable) says visual and auditory deficits can appear before cognitive decline in people with AD, and “Pathological changes in visual cortex” describes early changes in visual association areas. The paper does not establish the claimed link for each named region, particularly MGD. The eleven ROI correlations are attributed to the citing paper’s Supplementary Table 10, not to this source. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [74] Albers, M. W. et al. At the interface of sensory and motor dysfunctions and Alzheimer’s disease. Alzheimers Dement. 11, 70–98 (2015). source

      The Visual System neuropathology section reports that visual cortical areas can be affected early in AD, supporting the visual-region portion. The Auditory System neuropathology section reports pathology in the medial geniculate nucleus but says its presence in presymptomatic patients with hearing loss has not been investigated, so it does not establish that region’s link to early human symptoms. The eleven-ROI correlations are the citing paper’s own result, not a finding attributed to this source. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

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

    Premisecitation

    PFT and AVT both have roles in cognition and memory.

    “and both have roles in cognition and memory67.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [67] Aggleton, J. P., Pralus, A., Nelson, A. J. & Hornberger, M. Thalamic pathology and memory loss in early Alzheimer’s disease: moving the focus from the medial temporal lobe to Papez circuit. Brain 139, 1877–1890 (2016). source

      The Introduction (page unavailable) links the anterior thalamic nuclei to cognition and episodic memory, and Clinical and therapeutic implications (page unavailable) describes roles in attention and learning. Our copy does not identify PFT or AVT by those abbreviations or establish the claim separately for both, so the specific two-structure attribution is not fully supported. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    All five parental strains have been deeply sequenced and all F1 hybrids therefore have completely defined genomes.

    “All five of the parental strains have been deeply sequenced and all F1 hybrids therefore have completely defined genomes11,82.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [11] Ashbrook, D. G. et al. A platform for experimental precision medicine: the extended BXD mouse family. Cell Syst. 12, 235–247 (2021). source

      The Introduction reports full sequence data for both BXD parents, not five parental strains. The Discussion (Future Directions) says that isogenic BXD F1 lines have entirely defined genomes, supporting the genome claim for those hybrids but not the stated five-parent basis. (LLM, conf=0.90)

      Experimental — accuracy not yet measured
    • cites [82] Oakley, H. et al. Intraneuronal β-amyloid aggregates, neurodegeneration, and neuron loss in transgenic mice with five familial Alzheimer’s disease mutations: potential factors in amyloid plaque formation. J. Neurosci. 26, 10129–10140 (2006). source

      Materials and Methods says the engineered APP and PS1 transgenes were sequenced and describes breeding with B6/SJL F1 hybrids. Neither that passage nor the transgenic-strategy section addresses deep sequencing of five parental strains or whether the F1 hybrids have completely defined genomes. Searched for parental-strain sequencing and F1 genome definition. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

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

    Study design and conductcitation

    ROI delineation involved cycles of interactions between automated protocols and appraisal by skilled neuroanatomists following the strategy applied to human high throughput neuroimaging.

    “In brief, this has involved cycles of interactions between automated protocols described above and an appraisal by skilled neuroanatomists following the strategy applied to human high throughput neuroimaging95.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [95] Park, Y.-G. et al. Protection of tissue physicochemical properties using polyfunctional crosslinkers. Nat. Biotech. 37, 73–83 (2018). source

      The Materials and Methods sections describe automatic atlas alignment and human-reviewed, semi-automatic axon tracing, but not the claimed strategy for ROI delineation in human high-throughput neuroimaging. Searches for skilled neuroanatomist appraisal and iterative automated ROI delineation found no passage addressing that strategy. (LLM, conf=0.73)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Thalamic regions atrophy even in presymptomatic human carriers of PSEN1 mutations.

    “Thalamic regions atrophy even in presymptomatic human carriers of PSEN1 mutations70.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [70] Ryan, N. S. et al. Magnetic resonance imaging evidence for presymptomatic change in thalamus and caudate in familial Alzheimer’s disease. Brain 136, 1399–1414 (2013). source

  • May combine two claims.

    Study design and conductcitation

    The numerical analyses are relatively robust, and in both cases and controls the counts align reasonably well with well validated estimates of neuron numbers.

    “Nonetheless, the numerical analyses are relatively robust, and in both cases and controls our counts align reasonably well with well validated estimates of neuron numbers23.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [23] Tian, Y., Johnson, G. A., Williams, R. W. & White, L. A rapid workflow for neuron counting in combined light sheet microscopy and magnetic resonance histology. Front. Neurosci. 17, 1223226 (2023). source

      The Abstract (page unavailable) says the paper’s mouse neuronal-density estimates fall within the range of previous stereological studies. The Discussion (page unavailable) says most cortical-region densities were consistent with earlier reports, but identifies several regions with higher estimates. The paper does not establish that the citing authors’ counts align with validated neuron-number estimates in both cases and controls; its comparison is primarily of density in mice. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    Welch's t-tests did not detect any difference between male and female controls (P = 0.32), nor between all cases and all controls (P = 0.62).

    “Welch’s t-tests did not detect any difference between male and female controls (P = 0.32), but surprisingly, there were also no differences between all cases and all controls (P = 0.62, combined sexes).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    When F1s have a BXD77 sire, 124 of the 137 regions have significant 5×FAD effects at nominal P <0.05; corresponding numbers for BXD101, BXD32, and BXD65 sires are 41, 27, and 26.

    “When F1s have a BXD77 sire, 124 of the 137 regions have significant 5×FAD effects at a nominal P <0 .05 (see green ROI tags in Fig. 4c). Corresponding numbers for BXD101, BXD32, and BXD65 sires are 41, 27, and 26.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The transgene effects trend in precisely the same direction on all four genetic backgrounds; only the hypoglossal nucleus (HGN) is an exception.

    “Of more biological importance, the transgene effects trend in precisely the same direction on all four genetic backgrounds (Fig. 4c, the four blue and pink columns labeled %). Only the hypoglossal nucleus (HGN) is an exception.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    At a fine-grained level, 31 ROIs have significant differences even with a stringent Bonferroni correction, mainly on the BXD77 background.

    “At this fine-grained level 31 ROIs have significant differences even with a stringent Bonferroni correction, but mainly on the BXD77 background”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Effects of the 5×FAD transgene are more visually striking in females than males, whether the polarity is an increase or a decrease in volume, indicating higher female variance.

    “Effects of the transgene are more visually striking in females than males, and this is true whether the polarities are an increase or a decrease in volume (Fig. 4d)—a sign of higher female variance (Fig. 4d).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Both sex-stratified tests were highly significant: female cases had increased variance (P = 0.0025) and male cases had more consistent responses and reduced variance (P = 0.0046).

    “However, both sex-stratified tests were highly significant: female cases with increased variance (P = 0.0025); male cases with more consistent responses and reduced variance (P = 0.0046).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Several ROIs have strong sex differences only in cases or only in controls; for example, subthalamus (SUT) in cases, and striatum-like amygdalar nuclei (SLA) and medial forebrain bundle (mbh) in controls.

    “Several ROIs have strong sex differences only in cases or only in controls. For example, the subthalamus (SUT) has a strong sex difference in cases (−logP of 4.2), while the striatum-like amygdalar nuclei (SLA) and the medial forebrain bundle (mbh) have strong differences only in controls.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The reduction in NeuN-positive cell density is −18.2 ± 5.0% in CA1 and −24.5 ± 2.6% in CA3 relative to control.

    “The reduction of terms of density amounts to −18.2 ± 5.0% in CA1 and −24.5 ± 2.6% in CA3 relative to control.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The nucleus accumbens gains volume in 5×FAD cases but the dorsal striatum does not.

    “The nucleus accumbens gains volume but not the dorsal striatum.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Almost all diencephalic, mesencephalic and hindbrain structures have smaller volumes in 5×FAD cases, as well as most fiber tracts.

    “In contrast, almost all diencephalic, mesencephalic and hindbrain structures have smaller volumes, as well as most fiber tracts”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For contextual fear memory (CFM), 9 ROIs have significant correlations at PBH < 0.05 and 47 ROIs at PBH < 0.1.

    “Corresponding numbers for CFM are 9 and 47 ROIs.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Sex accounts for approximately 7% of the variance in the overall model that includes all ROIs and 7% of variance of the 137 ROIs significantly affected by the transgene.

    “Sex accounts for approximately 7% of the variance in the overall model that includes all ROIs and sex also accounts for 7% of variance of the 137 ROIs significantly affected by the transgene.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Of the 137 ROIs significantly affected by the transgene, 123 are significant in females and 93 are significant in males.

    “Of these ROIs 123 are significant in females and 93 are significant in males (green bars in Fig. 4d).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Total ventricular volume is reduced in 5×FAD cases but the difference is insignificant.

    “The total ventricular volume (Vnt) is reduced, but the difference is insignificant.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The polarity of volume effects is reversed for all thalamic and brainstem structures in AD-BXD77 F1 cases, and this is highly significant (Fisher combined P <0.0001).

    “In contrast, the polarity of effects is reversed for all thalamic and brainstem structures (Fig. 6h) and are again highly significant (Fisher combined P of <0.0001).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Twenty-nine ROIs have significant correlations with contextual fear acquisition (CFA) at PBH < 0.05 and 56 ROIs at PBH < 0.1.

    “Twenty-nine ROIs have significant correlations with CFA at a PBH < 0.05 and 56 ROIs at a PBH < 0.1”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The majority of telencephalic grey matter structures have larger volumes in 5×FAD cases; only the striatum (STR) and pallidum (PAL) are exceptions.

    “The majority of telencephalic grey matter structures have larger volumes; only the striatum (STR) and pallidum (PAL) are exceptions.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The dorsal lateral geniculate nucleus (LGD) and inferior colliculus (ICC and ICO) are notable exceptions to the rostrocaudal gradient, with large volumes.

    “Two other notable exceptions are the dorsal lateral geniculate nucleus (LGD) and the inferior colliculus (ICC and ICO)—both of which have large volumes”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Genetic background explains 23% of the variance in fractional volume of all ROIs and 20% of the variance of the 137 ROIs significantly altered by the transgene.

    “The genetic background into which the transgene is placed—in our study defined by the strain of the BXD sire to which the C57BL/6J congenic dam is crossed—explains 23% of the variance in fractional volume of all ROIs, and 20% of that of the 137 ROIs with volumes significantly altered by the transgene when we test using a well-powered full model”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Correlations of fear memory and volumes of hilus of the dentate gyrus (DGH) and posterior amygdala (PAN) are both strongly negative (ρ = −0.48 for DGH and −0.49 for PAN).

    “Correlations of fear memory and volumes of hilus of the dentate gyrus (DGH) and the posterior amygdala (PAN) are both strongly negative (Fig. 7b,d). Three transgenic females have anomalously large volumes and poor memory retention, but even rank correlations are negative (ρ of −0.48 for DGH and −0.49 for PAN).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Fourteen ROIs are nominally significant for sex-by-transgene interaction effects, including subthalamus (SUT, P = 0.0016), pontine central gray (PCG, P = 0.011), and hippocampus (HPO, P = 0.018); none survive multiple test correction.

    “Fourteen ROIs are nominally significant including the subthalamus (SUT, P = 0.0016), the pontine central gray (PCG, P = 0.011), and the hippocampus (HPO, P = 0.018). While intriguing, none of these interactions stand up to a multiple test correction.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The CA1 reduction in NeuN-positive cell density is not significant (PBH = 0.39), but the CA3 reduction is significant (PBH = 0.04).

    “In the case of CA1 this difference is not significant (PBH = 0.39) but is in the case of CA3 (PBH = 0.04).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The nine significant ROIs for CFM are TPA, THA, LAT, LAA, CTS, DGH, BLA, DGT, and PAN-a list enriched for thalamic, amygdalar, and hippocampal ROIs.

    “In comparison, the nine significant ROIs for CFM are TPA, THA, LAT, LAA, CTS, DGH, BLA, DGT, and PAN—a list that is enriched for thalamic, amygdalar, and hippocampal ROIs.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The link between increased hippocampal volume and neuron loss is most clearcut in CA3 but is also likely true of CA1.

    “This is most clearcut in CA3 but is also likely to be true of CA1 (Fig. 6i,k).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Fourteen of the 29 ROIs are thalamic nuclei (AVT, ILM, PCN, PFT, PIT, POT, SU, THA, TPA, TSM, VAL, VGT, VMT, and ZIN), as well as the lateral forebrain bundle (lbu); two others are tracts associated with the hippocampus (for and fim); one is the basomedial amygdala (BMA); and the remaining seven are motor-related nuclei and tracts (ION, MRN, MBM, PNM, PRT, RED, TMN, and cst).

    “Fourteen of these are thalamic nuclei: the AVT, ILM, PCN, PFT, PIT, POT, SU, THA, TPA, TSM, VAL, VGT, VMT, and ZIN, as well as the lateral forebrain bundle (lbu). Two others are tracts associated with the hippocampus (for and fim); one is the basomedial amygdala (BMA); and the remaining seven are motor-related nuclei and tracts (ION, MRN, MBM, PNM, PRT, RED, TMN, and cst).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Forty-six ROIs with opposite polarities of correlations between cases and controls are nominally significant with fear acquisition, but none meet an FDR correction at PBH < 0.05.

    “Although 46 ROIs with opposite polarities of correlations between cases and controls are nominally significant with fear acquisition (e.g., Fig. 7c,e), none meet an FDR corrections at PBH < 0.05.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The only midbrain structure that expands significantly is the inferior colliculus (ICO), a key region critical for auditory processing.

    “Similarly, the only midbrain structure that expands significant is the inferior colliculus (Fig. 4a,b, ICO)—a key region critical for auditory processing.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Both temporal association cortex (TAC) and paraventricular nucleus of the thalamus (PVT) differ markedly in correlations between cases and controls for CFA (PVT rho is −0.38 for cases and +0.44 for controls).

    “Both temporal association cortex (TAC) and the paraventricular nucleus of the thalamus (PVT) also differ markedly (Fig. 7c, where rhois −0.38 for cases and +0.44 for controls).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The loss of volume of the corticofugal projects in the cst strongly supports the finding of covariation, as does the 11% loss of FA signal in this tract.

    “The loss of volume of the corticofugal projects in the cst that is mainly made up of large descending layer 5 axons (Fig. 6g, cst) strongly supports this finding, as does the 11% loss of FA signal in this vital tract (Fig. 6j).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    A subset of 56 ROIs has reversed polarities of correlations of fear acquisition and volume between cases and controls, but only 13 ROIs for fear memory.

    “A subset of 56 ROIs has reversed polarities of correlations of fear acquisition and volume between cases and controls (e.g., Fig. 7c,e) but only 13 ROIs for fear memory.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingnone given

    The volume reductions in diencephalic, midbrain, and hindbrain ROIs are just as strong and just as specific as the volume expansions in most telencephalic regions.

    “The volume reductions in diencephalic, midbrain, and hindbrain ROIs are just as strong and just as specific as the volume expansions in most telencephalic regions.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    The authors reviewed delineations of VCS and recounted all cases and controls and could not identify any operational errors.

    “We have reviewed the delineations of VCS and recounted all cases and controls, and we could not identify any operational errors.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    Regions that gain volume correspond roughly to Swanson's dorsolateral subsystem that mediates voluntary behaviors and cognition, whereas regions that lose volume correspond to the ventromedial subsystem that modulates affective and homeostatic functions.

    “Regions that gain volume correspond roughly to Swanson’s dorsolateral subsystem that mediates voluntary behaviors and cognition, whereas regions that lose volume correspond to the ventromedial subsystem that modulates affective and homeostatic functions45.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [45] Sokoloff, L. et al. The [14C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J. Neurochem. 28, 897–916 (1977). source

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

    Premisecitation

    Cortical areas are 250 μm thicker in presymptomatic carriers of PSEN1 mutations than in controls at early prodromal stages, but are 250 μm thinner at late stages of disease.

    “particularly the finding that cortical areas are 250 μm thicker in presymptomatic carriers of PSEN1 mutations than in controls at early prodromal stages, but are 250 μm thinner at late stages of disease (see figure 5 of Ref46).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [46] Schwartz, W. J. & Gainer, H. Suprachiasmatic nucleus: use of 14C-labeled deoxyglucose uptake as a functional marker. Science 197, 1089–1091 (1977). source

  • May combine two claims.

    Interpretationcitation

    The rostro-caudal difference may reflect cortical swelling in the prodromal phase that is also detected in human carriers of aggressive PSEN1 mutations.

    “This rostro-caudal difference may reflect cortical swelling in the prodromal phase that is also detected in human carriers of aggressive PSEN1 mutations46.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [46] Schwartz, W. J. & Gainer, H. Suprachiasmatic nucleus: use of 14C-labeled deoxyglucose uptake as a functional marker. Science 197, 1089–1091 (1977). source

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

    Interpretationreasoning

    The consistency of the apparent swelling of most cortical regions may reflect inflammatory responses downstream of many possible transgene effects, including perhaps elevated β amyloid levels.

    “The consistency of the apparent swelling of most cortical regions may reflect inflammatory responses downstream of many possible transgene effects, including perhaps elevated β amyloid levels.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    We manually checked consistency of MRH-driven delineations and compared our boundaries to conventional stereological resources and stereological data.

    “We have manually checked consistency of MRH-driven delineations (e.g., Figs 2, 3) and compared our boundaries to conventional stereological resources and stereological data.”
    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

    All experiments using live mice were carried out in accordance with AAALAC standards and NIH Guide, and the protocol was approved by IACUC.

    “All experiments using live mice were carried out at University of Tennessee Health Science Center in accordance with standards of the Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC), as well as recommendations of the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Institutional Animal Care and Use”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    The overall ANOVA model evaluates sources of variance of fractional volumes as a function of transgene status, genetic background, sex, and MRI scanner.

    “The overall ANOVA model used in most analyses (Fig. 4) evaluates sources of variance of fractional volumes (bilateral sums of ROIs) as a function of transgene status, genetic background, sex, and MRI scanner.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Distortions in light-sheet processed tissue and images were removed by aligning the NeuN (561 nm) volume to the original DWI volume of the same individual.

    “The inevitable distortions in light-sheet processed tissue and images were removed by aligning the NeuN (561 nm) volume to the original DWI volume of the same individual (Fig. 1h).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    The combination of high resolution and high delineation precision has increased statistical power and enabled us to detect volumetric changes previously hidden by measurement noise.

    “The combination of high resolution and high delineation precision has increased statistical power and enabled us to detect volumetric changes previously hidden by measurement noise.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductcitation

    CVs are biased downward by about 25% with a sample size of only two, but we did not apply this correction because the technical components of CVs are overestimated by biological sources of asymmetry.

    “Like standard deviations, CVs are biased downward by about 25% with a sample size of only two93, but we did not apply this correction because the technical components of CVs are overestimated by biological sources of asymmetry.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [93] Gurland, J. & Tripathi, R. C. A simple approximation for unbiased estimation of the standard deviation. Am. Stat. 25, 30–32 (1971). source

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

    Study design and conductcitation

    The percentage of time spent freezing during the CFM test was used as a measure of CFM, a measure known to be dependent on hippocampal function.

    “The percentage of time spent freezing during this trial was used as a measure of CFM, a measure known to be dependent on hippocampal function84.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [84] Chen, C., Kim, J. J., Thompson, R. F. & Tonegawa, S. Hippocampal lesions impair contextual fear conditioning in two strains of mice. Behav. Neurosci. 110, 1177–1180 (1996). source

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

    Premisenone given

    Mouse models have benefits by enabling tight control of genetics, age, and diet, and because the impact of treatments can be evaluated within two years or less.

    “Mouse models have obvious benefits by enabling tight control of genetics, age, and diet, and because the impact of treatments can be evaluated within two years or less.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Statistical significance thresholds were set at P of 0.01, 0.05, and 0.1, corrected for 22 tests of ROIs.

    “Estimates of statistical significance used thresholds at P of 0.01, 0.05, and 0.1, corrected for 22 tests of ROIs (Fig. 6f-l).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    To minimize bias from individual specimens, a pairwise minimum deformation template (MDT) was generated using a hybrid image for all individuals of a particular genetic background type.

    “To minimize bias from individual specimens, a pairwise minimum deformation template (MDT) was generated using a hybrid image for all individuals of a particular genetic background type (e.g., Fig. 3a,b).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    This registration step is critical due to the small voxel size and the strength of the gradients.

    “This step is critical due to the small voxel size and the strength of the gradients (up to 50 times stronger than standard clinical DTI).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Benjamini-Hochberg or Bonferroni corrections were used to define P values at different FDR thresholds.

    “Benjamini-Hochberg (BH) corrections34 or Bonferroni corrections were used to define P values at different FDRs thresholds (Fig. 3d, Fig. 4).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [34] Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Series B Stat. Methodol. 57, 289–300 (1995). source

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

    Study design and conductreasoning

    Female transgenic mice were always used as dams so that the mitochondrial genome of all F1 hybrid was constant, and to control for strain-specific differences in pre- and postnatal parent-of-origin effects.

    “Female transgenic mice were always used as dams so that the mitochondrial genome of all F1 hybrid was constant, and to control for strain-specific differences in pre- and postnatal parent-of-origin effects.”
    Quote in the paperSubstantial claim

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

    Study design and conductreasoning

    We formally can only claim good precision of delineation rather than accuracy.

    “Note that we formally can only claim good precision of delineation rather than accuracy.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The scanner was upgraded in 2023 to an MR Solutions console-a factor controlled in all statistical analyses.

    “The system was upgraded in 2023 to an MR Solutions console—a factor controlled in all statistical analyses.”
    Quote in the paperNot a substantial claim

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

    Study design and conductreasoning

    There is a strong correlation between aβ plaque levels and NeuN+ cell densities in most ROIs in 5×FAD cases that is probably caused by common penetration patterns of these two antibodies.

    “2. a strong correlation between aβ plaque levels and NeuN+ cell densities in most ROIs in 5×FAD cases that is probably caused by common penetration patterns of these two antibodies.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    There is an implausible increase in NeuN+ cells in visual cortex that is either due to sampling error or lack of neuronal specificity of antibody binding in 5×FAD cases.

    “There are two exceptions: 1. an implausible increase in NeuN+ cells in visual cortex that is either due to sampling error or lack of neuronal specificity of antibody binding in 5×FAD cases”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    Major fiber tracts, ventricles and blood vessels are easily resolved in the MRH 3D volumes.

    “Major fiber tracts, ventricles and blood vessels are easily resolved.”
    Quote in the paperNot a substantial claim
73 of 73

  • The paper relies on this without support.

    There is not much systematic case-control data on structural effects of Alzheimer's disease in either humans or mice, particularly on the integrity of subcortical regions as a function of AD mutation type, sex, age of onset of behavioral symptoms, environment, or treatment history.

    “Another problem is that there is not much systematic case-control data on structural effects of AD in either humans or mice—particularly on the integrity of subcortical regions as a function of AD mutation type, sex, age of onset of behavioral symptoms, environment, or treatment history.”

  • The paper relies on this without support.

    The 5×FAD transgene in mice models the structural and behavioral effects of Alzheimer's disease.

    “We mapped the global impact of mutations in APP and PSEN1 across 231 precisely delineated regions of interest (ROIs) by diffusion tensor MRI at 25 μm resolution using diverse 5×FAD AD-BXD hybrid male and female mice at 14 months.”

  • The paper relies on this without support.

    The increase in volumes of ROIs has not previously been reported in 5×FAD mice.

    “The increase in volumes of ROIs has not previously been reported in 5×FAD mice.”

  • The paper relies on this without support.

    The gradients in volumetric effects—rostral expansion and caudal shrinkage, lateral expansion and medial shrinkage—are novel.

    “The gradients in volumetric effects—rostral expansion and caudal shrinkage, lateral expansion and medial shrinkage—are also novel and leads to key questions about ontogenetic, molecular, and cellular processes of AD vulnerability.”

  • The paper relies on this without support.

    The DMBA coordinate space is a consensus for the C57BL/6J strain at stereotaxic precision.

    “The DMBA coordinate space is a consensus for the C57BL/6J strain at stereotaxic precision.”
5 of 5

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • ConsistentF(3, 53) = 2.8reported p < .05 · recomputed p = .049

    “F(3,53) = 2.8, p < .05”

1 of 1

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.

  • The list of seven motor-related ROIs contains eight

    Minor

    The discussion calls these the remaining seven regions or tracts, but names eight: ION, MRN, MBM, PNM, PRT, RED, TMN, and cst.

  • A reported negative correlation has a positive coefficient

    Minor

    The reported coefficient and its ±0.14 range are positive, contrary to the description of the correlation as negative.

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Levene, H. in Contributions to Probability and Statistics: Essays in Honor of Harold Hotelling (eds Orkin, I. & Hotelling, H.) 278–292 (Stanford University Press, 1960).

Cited by

Still being checked.

Authors
Tian, Y., Hornburg, K., Austin, W., Cook, J. J., Qi, Y., White, L. E., Killmar, J. T., Kaczorowski, C. C., Ashbrook, D. G., Johnson, G. A., & Williams, R. W.
Affiliations
  • Duke Medical Center (US)
  • University of Tennessee Health Science Center (US)
  • University of Michigan (US)
  • Duke University (US)
Corresponding author
  • Ashbrook, David G.
  • Johnson, G. Allan
  • Williams, Robert W.
ORCID
Venue
Nature Neuroscience
Publisher
Springer Science and Business Media LLC
Published
2026-02-26
Volume
29
Issue
4
Pages
1007-1020
ISSN
  • 1097-6256 (ISSN-L)
  • 1546-1726
DOI
10.1038/s41593-025-02199-4
PMC
PMC13185701
OpenAlex
W7131681600
Semantic Scholar
34ea792ea991f7127421b389d0b2e0e2affb63f2
Cited by
1
References
103
Citations by year
2026: 1
FWCI
3.93
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Neuroscience (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion)
Funders
  • National Institutes of Health (5r01ns120954-04, 5r01ag070913-05)
  • National Institute on Aging (R01AG057914)
Topics
  • Health Sciences › Medicine › Physiology › Alzheimer's disease research and treatments
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
Keywords
Alzheimer's disease, brain structure, brain volume changes, APP mutations, PSEN1 mutations, 5xFAD mice, diffusion tensor imaging, brainstem, white matter tracts, hippocampus, amygdala, neocortex, thalamus, sex differences, fear acquisition, contextual memory, individual differences in behavior
MeSH terms
Alzheimer Disease, Animals, Behavior, Animal, Brain, Brain Mapping, Disease Models, Animal, Fear, Female, Magnetic Resonance Imaging, Male, Mice, Transgenic, Amyloid beta-Protein Precursor, Mice, Presenilin-1, Diffusion Tensor Imaging

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 2026-10-24.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-24.

Cited sources are checked by gpt-6-sol.

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