Claims 51
From the paper Degradation of perineuronal nets in hippocampal CA2 explains the loss of social cognition memory in Alzheimer's disease · claims file · checked 2026-10-21: ready
- fig1g-3m-pnn-area-no-difference
At 3 M, CA2 PNN area coverage does not differ between 5XFAD and control mice.
At 3 M, CA2 PNNs were indistinguishable between 5XFAD and controls
Results 3.1, paragraph 2; Figure 1C and 1GHow it will be judged
A claim of absence, judged on the paper's test. PNN-AREA per slice (unit is the slice, as the legend's n), 5XFAD and control at 3, 6, 9 and 12 M, in a two-way ANOVA on genotype and age with Tukey's multiple comparisons (Figure 1 legend). Reproduced when the Tukey comparison of 5XFAD against control at 3 M has p >= 0.05; not reproduced when p < 0.05. The mean difference and its p are reported, under each threshold reading.
- fig1g-pnn-area-reduced-6-to-12m
CA2 PNN area coverage is significantly lower in 5XFAD than in age-matched control mice at 6 M, and stays lower at 9 and 12 M.
at 6 M, we observed a significant reduction in CA2 PNN area in the 5XFAD mice ... This decrease in the PNN coverage persisted in the 9 and 12 M old 5XFAD
Results 3.1, paragraph 2; Figure 1D to 1GHow it will be judged
PNN-AREA per slice, two-way ANOVA on genotype and age with Tukey's multiple comparisons (Figure 1 legend). Reproduced when, at each of 6, 9 and 12 M, 5XFAD is below control with Tukey p < 0.05 under every threshold reading; partially reproduced when this holds at one or two of the three ages. The legend lists levels from p < 0.05 to p < 0.0001 without the figure, so p < 0.05 is the bound recorded.
- fig1h-wfa-intensity-reduced-6-to-12m
WFA intensity in CA2 is significantly lower in 5XFAD than in control mice at 6 to 12 M.
we computed WFA intensity in the CA2 area which was similarly reduced significantly at 6-12 M in 5XFAD mice
Results 3.1, paragraph 2; Figure 1HHow it will be judged
PNN-INTENSITY per slice, two-way ANOVA on genotype and age with Tukey's multiple comparisons (Figure 1 legend), under each intensity and threshold reading. Reproduced when, under every reading, 5XFAD is below control with Tukey p < 0.05 at each of 6, 9 and 12 M; partially reproduced when at one or two of these ages. If the readings give different outcomes the claim is methods_underspecified, with every result shown. The 3 M comparison is reported as well.
- fig1gh-aging-reduction-smaller-than-5xfad
Normal aging also reduces CA2 PNN coverage in control mice, but the loss is significantly more pronounced in 5XFAD mice.
Note that normal aging also presented with a reduction in PNN coverage, yet the loss of PNNs was significantly more pronounced in 5XFAD mice
Results 3.1, paragraph 2; Figure 1G and 1HHow it will be judged
PNN-AREA per slice, two-way ANOVA on genotype and age with Tukey's multiple comparisons over all eight cells (Figure 1 legend). The paper does not say which comparisons show the aging effect, so the reading of the reported finding is any older control group (6, 9 or 12 M) below control at 3 M with Tukey p < 0.05. Nor does it say which result makes the 5XFAD loss "significantly more pronounced"; the two-way ANOVA named in the legend tests this as the genotype by age interaction, which is used. Reproduced when the aging decline holds, the interaction has p < 0.05, and at each of 6, 9 and 12 M the 5XFAD loss relative to 5XFAD at 3 M is larger than the control's loss relative to control at 3 M. Partially reproduced when the aging decline holds and the 5XFAD loss is larger but the interaction is not significant. The same is reported for PNN-INTENSITY.
- fig1a-pv-pnns-intense-in-5xfad
In 6 M 5XFAD mice, the PNNs around PV neurons in CA2 remain relatively intense, while those around CA2 pyramidal neurons are lost.
however we observed relatively intense PV neuron-associated PNNs in 5XFAD mice
Results 3.1, paragraph 1; Figure 1A to 1DHow it will be judged
The paper reports no measure or test for this observation, and says the PNNs of pyramidal and PV neurons in CA2 could not be analysed separately (Results 3.1, paragraph 1). No source fixes how a PV neuron's PNN would be measured, and that choice could change the outcome, so the claim is blocked as methods_underspecified unless the dataset itself holds a per-PV-neuron WFA measure, which is then compared between 6 M 5XFAD and control by the Methods 2.10 default for two groups (two-tailed unpaired t-test); reproduced when the 5XFAD mean is not below the control mean at p < 0.05 while claim fig1g-pnn-area-reduced-6-to-12m holds at 6 M. As variants only, the mean WFA intensity in (a) a 2 um ring outside the PV cell outline and (b) the WFA mask overlapping the PV cell outline may be shown; they are not the basis of the outcome.
- fig1b-loss-of-wfa-peaks
Along the periphery of a single CA2 pyramidal neuron, the WFA line profile shows many high-intensity peaks in control and loses this peak and valley pattern in 5XFAD.
showing many high-intensity WFA peaks in control (dark magenta line) compared to the 5XFAD (light magenta line) group
Figure 1BHow it will be judged
The paper shows one example neuron per group and defines no peak measure, no peak threshold and no test. Testable only if the dataset holds the high magnification images or line profiles of Figure 1B, identified as such. Reproduced when the control profile has a higher maximum intensity than the 5XFAD profile and more peaks under peak prominence thresholds of both 10% and 25% of the control profile's intensity range; these two thresholds are not from the paper and stand for "many high-intensity peaks" only when they agree. Otherwise blocked as methods_underspecified.
- fig1i-pv-density-unchanged
PV neuron density in CA2 does not differ between 5XFAD and control mice at 3 or 6 M.
We did not find any change in either PV ... or NeuN cell density at 6 M of 5XFAD compared to age-matched controls
Results 3.1, paragraph 5; Figure 1IHow it will be judged
A claim of absence, judged on the paper's test. PV cells counted in ~20 um z-stacks divided by the PCP4-covered CA2 volume, per slice (Methods 2.8); 5XFAD against control by a two-tailed unpaired t-test at each age (Figure 1 legend). Reproduced when p >= 0.05 at both 3 and 6 M; not reproduced when either has p < 0.05. The difference and p are reported for each age.
- fig1j-neun-density-unchanged
NeuN neuron density in CA2 does not differ between 5XFAD and control mice at 3 or 6 M, so PNN loss precedes neuronal loss.
showing no difference between any groups
Results 3.1, paragraph 5; Figure 1JHow it will be judged
A claim of absence, judged on the paper's test. NeuN cells in single-plane images counted inside the PCP4-covered CA2 ROI and divided by its area, per slice (Methods 2.8); 5XFAD against control by a two-tailed unpaired t-test at each age (Figure 1 legend). Reproduced when p >= 0.05 at both 3 and 6 M; not reproduced when either has p < 0.05.
- figs2-pcp4-unchanged
PCP4 expression in CA2 is unaltered in 5XFAD mice at all ages studied, which is why it serves as the internal control for PNN area.
The expression of PCP4 remained unaltered in all studied age groups
Methods 2.8; Figure S2 (not in the text read)How it will be judged
A claim of absence. The Figure S2 legend was not read, so the paper's measure and test are not known. Readings the text allows, (a) PCP4 mask area in the CA2 rectangle per slice, two-way ANOVA on genotype and age with Tukey's comparisons as for Figure 1G, (b) the same measure, two-tailed unpaired t-test of 5XFAD against control at each age (Methods 2.10 default for two groups). Reproduced when no 5XFAD against control comparison has p < 0.05 under either reading; not reproduced when one does under both; if the readings disagree, methods_underspecified with both shown.
- figs3-other-regions-subtle
Outside CA2, the PNNs in other brain regions of 6 M 5XFAD mice show only subtle changes.
Although subtle changes in PNN expression were visible in some brain regions
Results 3.1, paragraph 1; Discussion, paragraph 2; Figure S3 (not in the text read)How it will be judged
The Figure S3 legend was not read, so its regions, measure and test are not known. Testable only if the dataset holds WFA images or values for regions outside CA2 at 6 M. With the Methods 2.8 measures and the Methods 2.10 default test (two-tailed unpaired t-test), reproduced when, in each region the dataset labels, the relative 5XFAD against control difference in PNN-INTENSITY is smaller in size than in CA2 of the same mice. Otherwise blocked.
- figs4-ca2-pnn-loss-j20-3xtg
CA2 PNNs are also conspicuously lost in two other AD models, J20 and 3XTg.
we assessed two more widely used AD models namely J20 and 3XTg, which also showed a conspicuous loss of PNN in CA2 area
Results 3.1, paragraph 1; Figure S4A to S4C (not in the text read)How it will be judged
The Figure S4 legend was not read, so the ages, sample sizes and test are not known. PNN-AREA and PNN-INTENSITY per slice, each model against its controls by a two-tailed unpaired t-test (Methods 2.10 default for two groups). Reproduced when both models are lower than their controls at p < 0.05 on at least one of the two measures; partially reproduced when one model is. Blocked if the dataset holds no J20 or 3XTg data.
- figs5-aggrecan-unchanged
Aggrecan expression in CA2 at 6 M is unaltered in 5XFAD mice, suggesting that the loss of WFA-labelled PNNs reflects a loss of chondroitin sulfate chains.
Aggrecan expression in CA2 at 6 M remains unaltered in 5XFAD mice
Results 3.1, paragraph 3; Figure S5A and S5B (not in the text read)How it will be judged
A claim of absence. The Figure S5 legend was not read. The text allows, for the measure, (a) aggrecan mask area / PCP4 area and (b) aggrecan intensity in the CA2 rectangle (the Methods 2.8 measures applied to the aggrecan channel), and for the test the Methods 2.10 default for two groups, a two-tailed unpaired t-test, on slices. Reproduced when 5XFAD against control has p >= 0.05 under both readings; not reproduced when p < 0.05 under both; methods_underspecified if they disagree.
- figs6-ca2-devoid-of-plaques
Despite widespread amyloid beta plaques in the hippocampus of 5XFAD mice, the CA2 area is largely devoid of plaques.
despite a widespread hippocampal deposition of amyloid beta plaques ... the CA2 area remained largely devoid of amyloid beta plaques
Results 3.1, paragraph 4; Figure S6A to S6C (not in the text read)How it will be judged
The Figure S6 legend was not read, so the measure is not known. Readings, (a) amyloid beta mask area fraction, (b) plaque count per area, in CA2 (the PCP4 rectangle) against the rest of the hippocampus in the same 5XFAD sections. Reproduced when, under both readings, CA2 is below the rest of the hippocampus in every 5XFAD section imaged with an amyloid stain; blocked if the dataset holds no amyloid stain.
- fig2b-3m-intact-social-memory
At 3 M, both 5XFAD and control mice have normal social memory, spending less time with the familiar mouse than with Novel 1 and more time with Novel 2.
We found that the 3 M old 5XFAD mice with intact PNNs behave identically to control mice, spending less time with the familiar mouse and more time with the novel mouse
Results 3.2, paragraph 3; Figure 2BHow it will be judged
SOCIAL and TRIALS-ANOVA with repeated measures on trial (the legend names it), under Tukey readings (i) and (ii). Reproduced when both groups show INTACT memory under every reading; partially reproduced when both comparisons go in the INTACT direction in both groups but only two or three of the four reach p < 0.05.
- fig2cd-3m-difference-scores-no-difference
At 3 M, the difference scores (Novel 1 - Familiar and Familiar - Novel 2) do not differ between 5XFAD and control mice.
with no significant difference in the difference scores between the groups
Results 3.2, paragraph 3; Figure 2C and 2DHow it will be judged
A claim of absence, judged on the paper's test. DIFF, two-tailed unpaired t-test for each score (n = 7 control, 9 5XFAD). Reproduced when both scores have p >= 0.05; not reproduced when either has p < 0.05. Group means and p are reported.
- fig2e-6m-impaired-social-memory
At 6 M, control mice show intact social memory, while 5XFAD mice show no difference in interaction time between Novel 1 and Familiar, nor between Familiar and Novel 2.
Specifically, unlike age-matched controls, 5XFAD mice showed no difference in investigation times between the first novel mouse and the familiar mouse, nor between the familiar mouse and the second novel mouse
Results 3.2, paragraph 3; Figure 2EHow it will be judged
SOCIAL and TRIALS-ANOVA with repeated measures on trial (legend), Tukey readings (i) and (ii). Reproduced when, under every reading, controls show INTACT memory and 5XFAD show IMPAIRED memory; partially reproduced when controls are INTACT and 5XFAD have one of the two comparisons at p < 0.05; not reproduced otherwise. The 5XFAD part is a claim of absence and is judged on this test as written.
- fig2fg-6m-difference-scores-lower
At 6 M, both difference scores show weaker discrimination of familiar and novel mice in 5XFAD than in control mice.
However, the lower difference scores in 5XFAD mice indicate a diminished ability to distinguish between familiar and novel mice
Results 3.2, paragraph 3; Figure 2F and 2GHow it will be judged
DIFF. Reproduced when both scores differ at p < 0.05 by the two-tailed unpaired t-test with the control score further from zero on the side of intact memory; partially reproduced when one does.
- figs7-nor-training-no-side-preference
During the training phase of the object recognition test, neither control nor 5XFAD mice prefer the right or the left object.
During the training phase both control and 5XFAD mice did not show any preference toward right or left object
Results 3.2, paragraph 4; Figure S7 (not in the text read)How it will be judged
A claim of absence. The Figure S7 legend was not read, so the paper's test is not known. Readings, on percent time with each object in the 10 min familiarization phase, (a) two-way mixed-effects ANOVA on genotype and side with Sidak's comparisons of left against right within genotype, as the paper used for the test phase (Figure 2I), (b) a two-tailed paired t-test of left against right within each genotype. Reproduced when no comparison has p < 0.05 under either reading; not reproduced when one does under both; methods_underspecified if they disagree.
- fig2i-nor-novel-object-preference
At 6 M, both control and 5XFAD mice spend significantly more time with the novel object than with the familiar one.
Bar graph showing significantly higher time spent with the novel object compared to familiar by both 6 M control and 5XFAD
Results 3.2, paragraph 4; Figure 2IHow it will be judged
Percent of object exploration time with the novel and the familiar object in the 10 min test phase (Methods 2.6.2), a two-way mixed-effects ANOVA on genotype and object (object within mouse) with Sidak's comparisons of novel against familiar within each genotype (legend). Reproduced when novel exceeds familiar with p < 0.05 in both genotypes; partially reproduced when in one.
- fig2j-nor-discrimination-index-no-difference
The novel object discrimination index does not differ between 6 M control and 5XFAD mice.
with no significant differences in the discrimination index between the groups
Results 3.2, paragraph 4; Figure 2JHow it will be judged
A claim of absence, judged on the paper's test. Discrimination index (novel - familiar) / (novel + familiar) exploration time in the test phase (Methods 2.6.2), two-tailed unpaired t-test (n = 11 control, 17 5XFAD). Reproduced when p >= 0.05; not reproduced when p < 0.05. Means and p are reported.
- fig2k-distance-moved-no-difference
Total distance moved does not differ between 6 M control and 5XFAD mice, so the social memory deficit is not due to a change in locomotion.
We measured the total travel distance in 6-M-old 5XFAD and control mice and found no significant difference between these groups
Results 3.2, paragraph 5; Figure 2KHow it will be judged
A claim of absence, judged on the paper's test, a two-tailed unpaired t-test (n = 17 control, 22 5XFAD). Open, in which session distance was measured, as the text does not say and n differs from Figure 2J; readings (a) the 10 min habituation in the open field, (b) the 10 min familiarization phase, (c) the 10 min test phase, each as the dataset holds it. Reproduced when p >= 0.05 under the readings the dataset allows; not reproduced when p < 0.05 under all; methods_underspecified if they disagree.
- figs8-baseline-social-memory-before-injection
Before AAV Cre injection, Acan fl/fl and WT mice both have normal social memory.
We began by confirming the normal social behavior of Acan fl/fl and WT mice
Results 3.3, paragraph 1; Figure S8 (not in the text read)How it will be judged
SOCIAL and TRIALS-ANOVA, Figure S8 legend not read, so readings (a) and (b) and (i) and (ii) all apply. Reproduced when both genotypes show INTACT memory under every reading; partially reproduced when one does.
- fig3b-acan-ko-eliminates-ca2-pnns
Eight weeks after CA2-targeted AAV SynCre injection, the vast majority of WFA-labelled PNNs in CA2 of Acan fl/fl mice are eliminated.
the vast majority of WFA-expressing PNNs were eliminated after 8 weeks of AAVCre injection in Acan fl/fl mice
Results 3.3, paragraph 1; Figure 3A and 3BHow it will be judged
The paper shows images only and reports no test. With the Methods 2.8 measure and the Methods 2.10 default for two groups, PNN-AREA in CA2 of Acan AAV Cre against WT AAV Cre mice by a two-tailed unpaired t-test. Open, the unit (no n is given); readings (a) slices, (b) mice, slices averaged per mouse. Reproduced when, under every reading, Acan AAV Cre is lower at p < 0.05 and its mean is less than half the WT AAV Cre mean ("vast majority").
- figs9-partial-knockout
A minor population of CA2 neurons in Acan AAV Cre mice shows only a modest reduction of aggrecan and WFA intensity, a partial knockout.
However, a minor population of CA2 neurons showed modest reduction in Acan and WFA fluorescence intensities
Results 3.3, paragraph 1; Figure S9A to S9C (not in the text read)How it will be judged
The Figure S9 legend was not read and the text defines no per-neuron measure and no cut-off for a "modest reduction". Testable only if the dataset holds per-neuron aggrecan and WFA intensities with the paper's own classification of the partially knocked out neurons; then reproduced when those neurons are fewer than half of the PCP4 neurons and their mean aggrecan and WFA intensities lie between those of the other Acan AAV Cre neurons and the WT AAV Cre neurons. Otherwise blocked as methods_underspecified. As a variant only, the share of Acan AAV Cre PCP4 neurons whose aggrecan and WFA intensities are above the lowest WT AAV Cre neuron may be shown.
- figs9-wt-cre-pnns-intact
WT mice injected with AAV SynCre keep unaltered CA2 PNNs.
WT mice injected with AAV SynCre displayed unaltered PNNs
Results 3.3, paragraph 1; Figure S9D (not in the text read)How it will be judged
A claim of absence. Figure S9D's legend was not read, so the comparison group is not known; the reading the text allows is PNN-AREA in CA2 of WT AAV Cre mice against uninjected WT or control mice of similar age held in the dataset, by a two-tailed unpaired t-test (Methods 2.10 default). Reproduced when p >= 0.05; blocked if no comparison group is in the dataset.
- fig3c-acan-ko-impairs-social-memory
After CA2 aggrecan deletion, Acan fl/fl mice show impaired social memory while WT mice injected with AAV SynCre show normal social memory.
By contrast, Acan fl/fl mice injected with AAV SynCre showed impaired social memory as evidenced by no difference in interaction times between familiar and novel mice
Results 3.3, paragraph 1; Figure 3CHow it will be judged
SOCIAL and TRIALS-ANOVA, readings (a), (b), (i), (ii). Reproduced when, under every reading, WT AAV Cre shows INTACT memory and Acan AAV Cre shows IMPAIRED memory; partially reproduced when WT AAV Cre is INTACT and Acan AAV Cre has one of the two comparisons at p < 0.05.
- fig3de-acan-ko-difference-scores
Both difference scores are significantly reduced in CA2 PNN-deleted mice (Acan AAV Cre) compared with WT AAV Cre mice.
The difference score between the two groups was significantly different
Results 3.3, paragraph 1; Figure 3D and 3EHow it will be judged
DIFF. Reproduced when both scores differ at p < 0.05 with the WT AAV Cre score further from zero on the side of intact memory; partially reproduced when one does.
- figs1-chabc-removes-wfa-adds-2b6
ChABC injection removes WFA staining in CA2 while the chondroitin sulfate stub epitope 2B6 appears; cortex near the injection site also shows low WFA intensity, while cortex distal to it is unchanged.
PNN disruption was confirmed by the loss of Wisteria floribunda agglutinin (WFA) staining in CA2 ... in combination with the appearance of CS stub 2B6 staining ... Cortical areas proximal to the injection site also showed a low WFA intensity ... however, cortical areas distal to the injection site showed no change in WFA intensity
Methods 2.2.1; Figure S1A to S1E (not in the text read)How it will be judged
The Figure S1 legend was not read, so the tests and n are not known. WFA intensity (Methods 2.8, PNN-INTENSITY readings) and 2B6 intensity in CA2, and WFA intensity in proximal and distal cortex, ChABC against PNase mice by two-tailed unpaired t-tests (Methods 2.10 default). Reproduced when CA2 WFA is lower and CA2 2B6 higher in ChABC at p < 0.05, proximal cortex WFA lower at p < 0.05, and distal cortex WFA has p >= 0.05; partially reproduced when the CA2 parts hold and a cortical part does not. Blocked when the dataset holds no 2B6 or cortical data, for those parts.
- figs1-heat-inactivated-chabc-no-effect
Heat-inactivated ChABC, like PNase, does not disrupt PNNs.
we compared the effect of heat-inactivated ChABC with the PNase injected control, both of which showed no disruption of PNNs
Methods 2.2.1; Results 3.3, paragraph 4; Figure S1 (not in the text read)How it will be judged
A claim of absence. Test not known (legend not read); the reading the Methods give for two groups is a two-tailed unpaired t-test of PNN-AREA or PNN-INTENSITY in CA2, heat-inactivated ChABC against PNase. Reproduced when p >= 0.05 on both measures; not reproduced when p < 0.05 on either with heat-inactivated ChABC lower.
- fig3g-chabc-pnn-loss-and-regrowth
ChABC disrupts CA2 PNNs at 5 days post injection, and PNNs reappear by 14 days.
At 5 DPI, ChABC-treated mice showed disrupted CA2 PNNs ... By 14 DPI, PNNs reappeared in the CA2 area
Results 3.3, paragraphs 3 and 4; Figure 3GHow it will be judged
The paper shows images only and reports no test. PNN-AREA in CA2, ChABC against PNase at 5 and at 14 DPI, two-tailed unpaired t-tests (Methods 2.10 default). Open, the unit; readings (a) slices, (b) mice. Reproduced when, under every reading, ChABC is lower at 5 DPI with p < 0.05 and its 14 DPI mean is above its 5 DPI mean; the 14 DPI comparison with PNase is reported.
- fig3h-5dpi-chabc-impairs-social-memory
At 5 days after ChABC injection into CA2, social memory is impaired, while PNase-injected control mice show normal social memory.
Graph showing impaired social memory in ChABC-injected mice at 5 DPI (5DPI ChABC) whereas PNase-injected mice (control) show normal social memory
Results 3.3, paragraph 4; Figure 3HHow it will be judged
SOCIAL and TRIALS-ANOVA, readings (a), (b), (i), (ii). Reproduced when, under every reading, PNase mice show INTACT memory and ChABC mice IMPAIRED memory; partially reproduced when PNase is INTACT and ChABC has one of the two comparisons at p < 0.05.
- fig3ij-5dpi-difference-scores
At 5 DPI, both difference scores of ChABC mice are significantly lower than those of PNase mice.
the difference score of the ChABC-treated group was significantly lower than the control at 5DPI
Results 3.3, paragraph 4; Figure 3I and 3JHow it will be judged
DIFF. Reproduced when both scores differ at p < 0.05 with the PNase score further from zero on the side of intact memory; partially reproduced when one does.
- fig3k-14dpi-social-memory-restored
By 14 days after ChABC injection, when PNNs have regrown, social memory is reinstated, with normal interaction behavior like PNase mice.
the social memory was also reinstated in ChABC-treated mice by 14DPI, as they exhibited normal social interaction behavior with familiar and novel mice, similar to PNase-treated control mice
Results 3.3, paragraph 4; Figure 3KHow it will be judged
SOCIAL and TRIALS-ANOVA, readings (a), (b), (i), (ii). Reproduced when, under every reading, both PNase and 14 DPI ChABC mice show INTACT memory; partially reproduced when ChABC shows one of the two comparisons at p < 0.05 in the INTACT direction.
- fig3lm-14dpi-difference-scores-no-difference
At 14 DPI, the difference scores of ChABC and PNase mice do not differ.
No difference was observed in the difference scores between the groups
Results 3.3, paragraph 4; Figure 3L and 3MHow it will be judged
A claim of absence, judged on the paper's test. DIFF, two-tailed unpaired t-test for each score (n = 8 PNase, 10 ChABC). Reproduced when both have p >= 0.05; not reproduced when either has p < 0.05.
- fig4b-3m-degs
In CA2 of 3 M 5XFAD mice compared with controls, 98 genes are upregulated and 133 downregulated.
In 3 M old 5XFAD mice compared to control, we found 98 upregulated genes and 133 downregulated genes
Results 3.4, paragraph 1; Figure 4BHow it will be judged
From the gene read counts, DESeq2 (Methods 2.5), genes with adjusted p < 0.05 and absolute log2 fold change > 1.00 (Methods 2.5; the Results print "log2FC < -1.00" here and "> 1.00" for 7 M, and the Methods' absolute threshold is followed). Adjusted p is DESeq2's default Benjamini-Hochberg with its default independent filtering (the Methods name no other). Open, the model; readings (a) DESeq2 on the 3 M samples alone with genotype as the factor, (b) one DESeq2 model on all samples with a four-level group factor (age and genotype) and the 3 M contrast, each (c) with unshrunken log2 fold changes (results() default) or (d) with lfcShrink (apeglm, the DESeq2 1.44 default type). Reproduced when both counts are within tolerance under every reading; partially reproduced when the direction balance holds (more down than up) and one count is within tolerance.
- fig4c-7m-degs
In CA2 of 7 M 5XFAD mice compared with controls, 314 genes are upregulated and 6 downregulated.
In 7 M old 5XFAD mice compared to control, there were 314 upregulated genes and 6 downregulated genes
Results 3.4, paragraph 1; Figure 4CHow it will be judged
As fig4b-3m-degs, for the 7 M contrast, under the same readings. Reproduced when both counts are within tolerance under every reading; partially reproduced when the up count is within tolerance and the down count is below 20 (most changes up).
- fig4d-3m-go-memory-terms-down
Among the top 10 GO terms at 3 M, many pathways of learning, memory and cognition are downregulated in 5XFAD mice.
we found many pathways associated with learning, memory, and cognition are downregulated in 3 M old 5XFAD mice compared to their age-matched controls
Results 3.4, paragraph 2; Figure 4DHow it will be judged
The DEGs of fig4b-3m-degs; GO terms ranked by adjusted p, top 10 (Figure 4D legend), each with the GOplot z-score (up - down) / sqrt(count) of its DEGs (GOplot, Methods 2.5). Open, the enrichment method, ontology and background, as the paper names only GOplot for plotting; readings (a) over-representation of the DEGs in GO Biological Process against all genes tested by DESeq2, (b) the same against all annotated mouse genes, both with Benjamini-Hochberg adjustment. Reproduced when, under every reading, at least three of the top 10 terms name learning, memory or cognition and have negative z-scores; partially reproduced for one or two such terms.
- fig4e-7m-go-immune-terms-up
Among the top 10 GO terms at 7 M, many immune response pathways are upregulated in 5XFAD mice.
Whereas many pathways related to immune response were upregulated in 7 M old 5XFAD mice compared to their age-matched controls
Results 3.4, paragraph 2; Figure 4EHow it will be judged
As fig4d-3m-go-memory-terms-down, on the DEGs of fig4c-7m-degs, readings (a) and (b). Reproduced when at least three of the top 10 terms name immune, inflammatory or defense responses, or immune cell activation, and have positive z-scores; partially reproduced for one or two.
- fig4fg-ecm-go-terms
ECM organization, assembly, disassembly and secretion terms change in 7 M 5XFAD against 3 M 5XFAD and against 7 M controls, but not in 3 M 5XFAD against 3 M controls.
GO term of the 3 M 5XFAD versus 7 M 5XFAD and 7 M 5XFAD versus 7 M control revealed changes in genes associated with the ECM organization, assembly, disassembly, and secretion ... Consistent with intact memory, 3 M 5XFAD versus 3 M control did not show any change in these genes
Results 3.4, paragraph 2; Figure 4F and 4GHow it will be judged
DEGs per contrast as in fig4b-3m-degs (with the 3 M against 7 M 5XFAD contrast added), GO enrichment under readings (a) and (b) of fig4d-3m-go-memory-terms-down. Reproduced when, under every reading, terms naming extracellular matrix organization, assembly, disassembly or secretion reach adjusted p < 0.05 in both the 7 M 5XFAD against 3 M 5XFAD and the 7 M 5XFAD against 7 M control contrasts, and none does in 3 M 5XFAD against 3 M control; partially reproduced when one of the two 7 M contrasts holds.
- fig4h-pnn-synthesis-genes
PNN synthesis genes are broadly stable; at 3 M, Vcan, Cspg5, Tnr and Hapln1 are downregulated in 5XFAD, and at 7 M, Vcan and Bgn are increased.
however, a few genes including Vcan, CSPG5, Tnr, and Hapln1 were downregulated ... Similarly, we found an increase in Vcan and Bgn components, while other genes remained mostly unaltered in 7 M old 5XFAD mice
Results 3.4, paragraph 3; Figure 4HHow it will be judged
The heatmap shows z-scored expression and the paper reports no test per gene. The direction is the sign of the DESeq2 log2 fold change of 5XFAD against control at each age (readings of fig4b-3m-degs); DESeq2 adjusted p is reported per gene. Reproduced when all four named genes are lower at 3 M and both named genes higher at 7 M under every reading; partially reproduced when at least four of the six go the named direction.
- fig4i-3m-remodeling-genes-up
At 3 M, 5XFAD CA2 shows increased expression of Mmp2, 3, 9, 14, 15 and 17, Timp1, 2 and 3, and Adam10, 11, 15 and 17.
3 M old 5XFAD compared to their age-matched controls shows an increase in MMP 2, 3, 9, 14, 15, 17, TIMP 1, 2, 3, and ADAM 10, 11, 15, 17
Results 3.4, paragraph 3; Figure 4IHow it will be judged
The sign of the DESeq2 log2 fold change of 5XFAD against control at 3 M for each of the 13 named genes (readings of fig4b-3m-degs); no test per gene is reported, and DESeq2 adjusted p is reported per gene. Reproduced when all 13 are positive under every reading; partially reproduced when at least 9 are.
- fig4i-7m-remodeling-genes-up
At 7 M, 5XFAD CA2 shows increased expression of Mmp2, 3 and 16, Timp1 and 3, and Adam15 and 17.
The 7 M old 5XFAD compared to their age-matched controls shows an increase in MMP 2, 3, 16, TIMP1, 3, and ADAM 15, 17
Results 3.4, paragraph 3; Figure 4IHow it will be judged
As fig4i-3m-remodeling-genes-up, for the seven named genes at 7 M. Reproduced when all 7 are positive; partially reproduced when at least 5 are.
- fig4i-common-and-7m-vs-3m-up
Mmp2, Mmp3 and Timp1 are upregulated at both 3 and 7 M, and 7 M 5XFAD mice show higher Mmp3, Timp1, 2 and 3, and Adam10 and 17 than 3 M 5XFAD mice.
The common genes upregulated at both 3 and 7 M 5XFAD compared to their age-matched controls were MMP2, MMP3, and TIMP1 ... whereas 7 M 5XFAD compared to 3 M 5XFAD mice show an increase in MMP3, TIMP1, 2, 3, Adam 10, 17
Results 3.4, paragraph 3; Figure 4IHow it will be judged
The sign of the DESeq2 log2 fold change (readings of fig4b-3m-degs) for Mmp2, Mmp3 and Timp1 in both age contrasts, and for the six named genes in the 7 M 5XFAD against 3 M 5XFAD contrast. Reproduced when all are positive under every reading; partially reproduced when at least two thirds are.
- fig4a-7m-social-memory-deficit
The 7 M 5XFAD mice used for RNA-seq have a confirmed social memory deficit, while social memory is still intact in the 3 M group.
We used 3 and 7 M old 5XFAD mice as comparative groups with confirmed social memory deficits at 7 M. The 3 M age represents the commencing phase of PNN disruption with still intact social memory
Results 3.4, paragraph 1; Figure 4AHow it will be judged
No figure shows these behavior results, and the paper gives no n or test for them (Table S4 not read). SOCIAL and TRIALS-ANOVA as in the other social memory claims, readings (a), (b), (i), (ii), on the 3 and 7 M control and 5XFAD mice of the RNA-seq cohort. Reproduced when, under every reading, 7 M controls show INTACT memory and 7 M 5XFAD mice show IMPAIRED memory; partially reproduced when 7 M controls are INTACT and 7 M 5XFAD have one of the two comparisons at p < 0.05. The 3 M groups are reported. Blocked if the dataset holds no behavior for this cohort.
- fig5-5m-pnns-normal
In 5XFAD mice, CA2 PNNs are normal until 5 M of age and impaired by 6 M.
we found that PNNs and social memory were normal until 5 M of age; however, by 6 M both are impaired
Results 3.5, paragraph 2How it will be judged
No figure shows PNNs at 5 M. PNN-AREA in CA2 at 5 M, 5XFAD against control, two-tailed unpaired t-test on slices (Methods 2.10 default; slices as in Figure 1). Reproduced when p >= 0.05 at 5 M and claim fig1g-pnn-area-reduced-6-to-12m holds at 6 M; blocked if the dataset holds no 5 M IHC.
- fig5b-5m-intact-social-memory
At 5 M, 5XFAD and control mice both show normal social memory.
when these mice exhibited normal social memory behavior with no significant difference in the difference score between 5 M old 5XFAD and age-matched controls
Results 3.5, paragraph 2; Figure 5BHow it will be judged
SOCIAL and TRIALS-ANOVA, readings (a), (b), (i), (ii). Reproduced when both groups show INTACT memory under every reading; partially reproduced when both comparisons go in the INTACT direction in both groups but only two or three reach p < 0.05.
- fig5cd-5m-difference-scores-no-difference
At 5 M, the difference scores do not differ between 5XFAD and control mice.
with no significant difference in the difference score between 5 M old 5XFAD and age-matched controls
Results 3.5, paragraph 2; Figure 5C and 5DHow it will be judged
A claim of absence, judged on the paper's test. DIFF, two-tailed unpaired t-test for each score (n = 7 control, 11 5XFAD). Reproduced when both have p >= 0.05; not reproduced when either has p < 0.05.
- fig5e-gm6001-preserves-social-memory
After a month of GM6001, 6 M 5XFAD mice show normal social memory, while vehicle-treated 5XFAD mice show disrupted social memory.
5XFAD mice treated with vehicle showed altered social memory behavior; however, the GM6001-treated 5XFAD mice exhibited normal social behavior
Results 3.5, paragraph 3; Figure 5EHow it will be judged
SOCIAL and TRIALS-ANOVA on the two 5XFAD groups (the Results say the behavior study included two groups; the legend's mention of age-matched controls gives no n), readings (a), (b), (i), (ii). Reproduced when, under every reading, GM6001 mice show INTACT memory and vehicle mice IMPAIRED memory; partially reproduced when GM6001 is INTACT and vehicle has one of the two comparisons at p < 0.05.
- fig5fg-gm6001-difference-scores
Both difference scores of GM6001-treated 5XFAD mice differ significantly from those of vehicle-treated 5XFAD mice.
with a significant difference in the difference score compared to vehicle-treated 5XFAD mice
Results 3.5, paragraph 3; Figure 5F and 5GHow it will be judged
DIFF. Reproduced when both scores differ at p < 0.05 with the GM6001 score further from zero on the side of intact memory; partially reproduced when one does.
- fig5i-gm6001-preserves-ca2-pnns
GM6001-treated 6 M 5XFAD mice have significantly higher CA2 PNN expression than vehicle-treated 5XFAD mice.
the IHC analysis of GM6001-treated mice brains exhibited significantly higher CA2 PNN expression compared to vehicle-treated 5XFAD mice
Results 3.5, paragraph 3; Figure 5H and 5IHow it will be judged
One value per mouse (the legend's n is mice), four groups, one-way ANOVA with Tukey's comparisons (legend). Open, what "PNN expression" is; readings (a) PNN-AREA, (b) PNN-INTENSITY (under its own readings). Open, how slices make a mouse's value; readings (1) the mean over the mouse's slices, (2) one slice per mouse if the dataset holds one. Reproduced when 5XFAD + GM6001 is above 5XFAD + vehicle with Tukey p < 0.05 under every reading; methods_underspecified if readings disagree, with all shown.
- fig5i-gm6001-controls-comparable
Control mice treated with GM6001 or vehicle have CA2 PNN expression comparable to each other and to GM6001-treated 5XFAD mice, so GM6001 has no significant effect in controls.
Age-matched controls treated with GM6001 or vehicle exhibited PNN expression comparable to GM6001-treated 5XFAD mice ... In addition, the control mice treated with GM6001 and vehicle showed comparable PNN expression
Results 3.5, paragraph 3; Figure 5IHow it will be judged
A claim of absence, judged on the paper's test, the one-way ANOVA with Tukey's comparisons of fig5i-gm6001-preserves-ca2-pnns, under its readings. Reproduced when control + GM6001 against control + vehicle, and each control group against 5XFAD + GM6001, all have p >= 0.05; not reproduced when any has p < 0.05. Differences and p are reported.
Pipeline
- Paper
- done (1 of 1)
- Claims
- done
- Claims check
- done (ready)
- Attempt
- todo
- Questions
- none
- Verified
- none