Corpus 1,272 assessed · 1,173 scored · 643 reproduced ≥75 · 168 flagged ·∅ 74.1/100
← New search

The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases.

Immunity · 2017
L1 67/100 3/4
Why this verdict

The main results reproduced, with only marginal, non-material deviations.

Reproduced on the brainbox compute brainarbeit.com
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Total score +10
✓ What held up
  • Same input data as the authors
What did not (or only partly)
  • 🟡Reported values were only indirectly comparable
  • 🟡A deviation arose in the data or preprocessing
  • 🟡A deviation was attributed to the published material
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The deviation was non-trivial in magnitude
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
67/100
Reproducibility score
0.4 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 29% of all assessed papers rank 795 of 1173 scored

A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.

Reproduction agent’s raw note

Reanalyzed all 6 publicly available GEO series underlying Krasemann et al. 2017 (Immunity, PMID 28930663) on «our HPC»/SLURM: 3 NanoString custom-chip series (GSE101686/7/8, parsed directly from RCC files) and 3 RNA-seq processed-matrix series (GSE102562/3/4). Using an independently constructed MGnD/homeostatic marker-panel score and per-gene Welch t-tests with BH-FDR correction (not the paper's exact limma/TMM/IPA/GSEA pipeline, and without raw FASTQ realignment -- out of scope per brief), four of the paper's core mechanistic claims reproduced with strong, statistically significant, directionally consistent evidence: (C1) Trem2-KO blocks MGnD induction in SOD1 mice; (C2) Clec7a-positive APP-PS1 microglia show a clear MGnD signature vs Clec7a-negative; (C3) Apoe-KO attenuates the phagocytosis-induced MGnD switch; (C5) EAE disease severity shows a clean dose-response increase in MGnD score. Two claims were only partially supported due to low replicate counts in the source data (C6: SOD1/APP-PS1 disease-progression trends, mostly n=1 per timepoint). One genuine, reproducible mismatch was found and is explicitly flagged rather than smoothed over: (C4) wild-type phagocytosis of dead neurons increases the MGnD score in the RNA-seq series (GSE102564, as the paper claims) but DECREASES it in both independent NanoString series (GSE101686 and GSE101688) that contain a comparable Phagocytic/NonPhagocytic contrast -- a cross-platform directional disagreement that could reflect a normalization-method artifact in this simplified reanalysis, small-n noise, or a genuine paradigm difference between cohorts; it was not adjudicated. All 6 datasets were complete, well-formed, and delivered what they promised (quality grade A across the board), though none of the GEO records state the paper's originally-reported sample size in a machine-parseable field, so n_reported is recorded as null (not guessed) throughout and only n_observed (from the deposited files) is asserted. Out of scope and not attempted: human post-mortem IHC/immunostaining figures, k-means clustering reproduction, Ingenuity Pathway Analysis / GSEA, limma batch-correction reproduction, and any raw FASTQ realignment -- these are wet-lab/manual or require tools/licenses beyond this room's scope.

These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.

✎ I am an author of this paper

Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.

Reason for the rerun

We email you a confirmation link first. The rerun is an objective re-measurement — it cannot change the verdict in your favour, only ask us to look again.

Provenance — full disclosure

When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.

Reproduced
2026-07-29
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-07-31
no human curator yet
Last updated
2026-07-31

Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.

Deep full-text extraction

Model: opus
Founding hypothesis

The paper asks which molecular mechanism drives the common switch of microglia from a homeostatic (M0) phenotype to a disease-associated neurodegenerative phenotype (MGnD) across neurodegenerative diseases, testing the hypothesis that a TREM2-induced APOE signaling pathway, triggered by phagocytosis of apoptotic neurons, mediates this switch.

Core claims
  • A common APOE-dependent microglial molecular signature (MGnD) — loss of homeostatic genes plus induction of inflammatory genes with Apoe among the most upregulated — occurs in ALS, MS and AD mouse models and around neuritic Aβ-plaques in human AD brain. finding
  • The TREM2-APOE pathway is a major regulator of the microglial functional phenotype in neurodegeneration and a target to restore homeostatic microglia. mechanism
  • Phagocytosis of apoptotic neurons switches microglia from the homeostatic to the MGnD phenotype, inducing Apoe and suppressing TGFβ-dependent homeostatic genes. mechanism
  • APOE acts cell-intrinsically in microglia to suppress homeostatic transcription factors (Mef2a, Mafb, Smad3, Egr1) and induce an inflammatory program (Bhlhe40, Tfec, Atf3, miR-155); miR-155 lies downstream of APOE. mechanism
  • TREM2 induces APOE signaling; genetic deletion of Trem2 suppresses Apoe and restores the homeostatic microglial signature in APP-PS1 and SOD1 mice. finding
  • Global or microglia-conditional deletion of Apoe, or deletion of Trem2, reduces neuronal loss in the acute facial nerve axotomy model of neurodegeneration. finding
  • The MGnD/MG-dNΦ signature is distinct from classical LPS/IFNγ-induced M1 microglia, in which Apoe is suppressed and Egr1 induced. finding
  • P2ry12 and Clec7a antibody staining discriminates M0-homeostatic from MGnD microglia and defines three microglial subsets relative to Aβ-plaques. method
Experimental setups
Assay System Perturbation Readout Platform
Nanostring gene expression profiling / transcriptome clustering (k-means) FACS-isolated brain and spinal cord microglia from SOD1 G93A (ALS), APP-PS1 (AD), EAE (MS) mice and aging mice disease models and aging (transgenic/immunization); none for controls microglial homeostatic and inflammatory gene expression during disease progression Nanostring
RNA sequencing FCRLS+ microglia subsets (Clec7a−, Clec7a int/lo, Clec7a+) from APP-PS1 mice; SOD1:Trem2−/− vs SOD1:Trem2+/− male and female microglia; WT vs Apoe−/− phagocytic and non-phagocytic microglia Trem2 knockout, Apoe knockout, apoptotic neuron injection genome-wide differential gene expression, MGnD signature genes, gender-specific gene clusters
Quantitative real-time PCR (qPCR) Sorted microglia from MOG-induced EAE in NOD (chronic-relapsing) and C57BL/6J (acute) mice; Cx3cr1CreERT2:Apoefl/fl vs Cx3cr1WT:Apoefl/fl microglia; miR-155−/− and Trem2−/− microglia EAE induction, tamoxifen-induced microglial Apoe deletion, Apoe/Trem2/miR-155 knockout, apoptotic neuron injection expression of Apoe, Clec7a, Csf1r, Tgfbr1, Tmem119, miR-155
Immunohistochemistry / immunofluorescence and quantitative image analysis APP-PS1 mouse brain, SOD1 mouse spinal cord, EAE mouse CNS, human AD post-mortem brain none (disease genotype); Trem2 knockout in APP-PS1 and SOD1 mice P2ry12, Clec7a, TMEM119, IBA1, APOE, Aβ-plaque and phosphorylated neurofilament (pNF) co-localization; Aβ-plaque load; microglial morphology
Intracortical/intrahippocampal injection of apoptotic neurons and FACS isolation of phagocytic vs non-phagocytic microglia Naïve WT, Apoe−/−, Trem2−/−, miR-155−/−, Cx3cr1CreERT2:Apoefl/fl mice injection of apoptotic neurons (dN), live neurons, apoptotic monocytes, Apoe−/− dN, E. coli or Zymosan particles, PBS, or recombinant APOE microglial phagocytosis, Apoe induction kinetics (3–16 h, restoration at 14 d), MGnD vs homeostatic gene expression
Quantitative mass spectrometry (proteomics) MG-dNΦ vs MG-nΦ microglia sorted from mouse brain apoptotic neuron injection protein abundance of Apoe, Lgals3, Rgs10, Bin1
Facial nerve axotomy (FNA) neuronal survival assay WT, Apoe−/−, Trem2−/−, Cx3cr1CreERT2:Apoefl/fl and Cx3cr1WT:Apoefl/fl mice facial nerve axotomy; global or microglia-conditional Apoe deletion; Trem2 knockout neuronal loss/survival in the axotomized facial motor nucleus
In vitro/in vivo phagocytosis blocking assay and in vivo M1 stimulation Mouse microglia with apoptotic neurons; LPS- and IFNγ-stimulated mouse microglia in vivo; kainic acid-injected mice Annexin V blockade of phosphatidylserine; LPS/IFNγ; kainic acid phagocytosis rate; homeostatic gene suppression; Apoe, Egr1, Arg1, Ym1, Il1b, Ptgs2, Ccl2, Ccl5, Tspo, Msr1, Cebpb expression; P2ry12−/Clec7a+ microglia at 48 h
Key results
  • Annexin V blockade of phosphatidylserine on apoptotic neurons reduced microglial phagocytosis 88%
  • Two gene clusters define the disease-associated signature: loss of 68 homeostatic microglial genes and upregulation of 28 inflammatory molecules including Apoe 68 genes down; 28 genes up
  • Mef2a, Sall1 and Tgfbr1 negatively correlated, and Apoe positively correlated, with disease progression in EAE, SOD1 and APP-PS1 models
  • In Apoe−/− phagocytic microglia, 885 genes induced in WT MG-dNΦ were repressed and 1,220 genes suppressed in WT MG-dNΦ were restored; 40 of 68 commonly disease-suppressed homeostatic genes were restored and 17 of the commonly upregulated disease genes were suppressed 885 repressed; 1,220 restored
  • APP-PS1:Trem2−/− microglia showed suppression of 7 inflammatory molecules (Trem2, Axl, Clec7a, Csf1, Itgax, Cd34, Apoe) and restoration of 108 genes, 54 of which were commonly suppressed homeostatic genes 7 down; 108 restored (54 homeostatic)
  • SOD1:Trem2−/− microglia showed 36 downregulated inflammatory genes (11 common to the disease signature) and 240 restored genes (66 commonly suppressed homeostatic genes); miR-155 was not induced 36 down; 240 restored
  • Global and microglia-conditional Apoe deletion, and Trem2 deletion, reduced neuronal loss after facial nerve axotomy
  • Loss of P2RY12+ microglia in human AD cortex correlated with axonal dystrophy but not with the extent of Aβ deposition; Aβ-plaque load was decreased in APP-PS1:Trem2−/− mice
Key statistics
  • count 68 homeostatic microglial genes lost; 28 inflammatory molecules upregulated (Cluster 1 and Cluster 2 of the common disease-associated microglia signature (Figure 1A))
  • fold_change reduced microglial phagocytosis by 88% (Annexin V blocking of phosphatidylserine on apoptotic neurons)
  • count 885 genes induced in WT MG-dNΦ repressed in Apoe−/− MG-dNΦ; 1,220 genes suppressed in WT MG-dNΦ restored in Apoe−/− (Clusters 1–3 of WT vs Apoe−/− phagocytic microglia transcriptomes)
  • count 17 of the commonly upregulated disease genes suppressed and 40 of 68 commonly suppressed homeostatic genes restored in Apoe−/− phagocytic microglia (Overlap with common disease signature)
  • count 7 inflammatory molecules suppressed; 108 restored genes, 54 commonly suppressed homeostatic genes (Nanostring profiling of APP-PS1:Trem2−/− brain microglia)
  • count 36 downregulated inflammatory genes (11 common to disease signature); 240 restored genes (66 homeostatic) (SOD1:Trem2−/− microglia)
  • count 279 commonly affected genes; 575 female-specific genes; 2,639 male-specific genes (RNAseq of SOD1:Trem2 microglia, gender-dependent clusters)
  • other Apoe increased as early as 3 h post-injection, peaked at 16 h; homeostatic P2ry12+Clec7a− microglia restored at 14 days; P2ry12−/Clec7a+ microglia at 48 h after kainic acid (Kinetics of MGnD induction after apoptotic neuron or kainic acid injection)

Statistical methods review

Model: sonnet

A neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.

The excerpt describes transcriptomic and pathway-level profiling of microglia across mouse models of ALS, AD, and MS and in human AD tissue, using k-means clustering to define gene modules, linear regression to relate gene expression to disease progression, Ingenuity Pathway Analysis (IPA) for upstream-regulator inference, and gene set enrichment analysis (GSEA) to compare signatures across models, alongside RNAseq, Nanostring, qPCR, and mass spectrometry profiling. The provided text does not include a dedicated statistics/methods section, so specific hypothesis tests, sample sizes, and error reporting for individual comparisons are not described here.

Replicationunclear GroupsDisease models (ALS/SOD1, AD/APP-PS1, MS/EAE) vs. controls, and genetic knockouts (Apoe−/−, Trem2−/−, Cx3cr1CreERT2:Apoefl/fl, miR-155−/−) vs. wild-type/control littermates, across ages and disease stages Pairingunclear Randomization/blindingnot stated Dispersionunclear
Statistical tests used
Test Applied to n Assumptions
k-means clustering Identification of homeostatic vs. neurodegenerative (MGnD) gene clusters from microglial transcriptomes (Figure 1A) not stated
Linear regression analysis Correlation of Mef2a, Sall1, Tgfbr1, and Apoe expression with disease progression in EAE, SOD1, and APP-PS1 models (Figure 1B) not stated
Ingenuity Pathway Analysis (IPA) upstream-regulator analysis Identification of APOE and TGFβ as upstream regulators of the MGnD signature, and of de-repressed molecules in Apoe−/− and miR-155−/− microglia (Figure 1D, Figure S6F) not stated
Gene set enrichment analysis (GSEA) Comparison of the MG-dNΦ phagocytic microglia signature against transcriptomes from aging, irradiation, AD, ALS, Mfp2-deficiency, neuropathic pain, and Mecp2-deficiency models (Figure 3I) not stated
Approaches that could also have been used
  • Disease-progression relationships between gene expression and clinical/disease stage were assessed with linear regression (Figure 1B).
    Could also: A mixed-effects or generalized additive model that accounts for repeated within-animal measurements over the disease course — Longitudinal or repeated-measures designs across disease stages often include correlated observations from the same animals over time; a mixed-effects framework can model that within-subject correlation explicitly alongside the trend.
  • Gene modules were defined using k-means clustering of microglial transcriptomes (Figure 1A).
    Could also: Hierarchical or consensus clustering with stability assessment — These approaches do not require pre-specifying the number of clusters and can provide an additional check on cluster robustness, which some readers find complementary to k-means results.
  • Pathway-level relationships (e.g., APOE and TGFβ as upstream regulators) were inferred using IPA's proprietary knowledge base (Figure 1D, S6F).
    Could also: An open-source enrichment approach such as GSEA/fgsea or gene ontology over-representation analysis with permutation-based false discovery rate control — Open, permutation-based enrichment tools allow independent reproduction of pathway calls and make the statistical model underlying enrichment scores fully transparent.
  • Many individual genes and multiple model systems (ALS, AD, MS, phagocytosis, knockouts) were compared in parallel across figures.
    Could also: A pre-specified multiple-comparison correction such as Benjamini-Hochberg FDR applied across the family of gene-level comparisons — When many genes or conditions are evaluated in parallel, an FDR or family-wise error correction is a standard way to characterize the expected proportion of false positives among the highlighted hits.
  • The excerpt does not specify how variability was summarized (e.g., SD, SEM, or CI) for the quantitative comparisons shown in figures.
    Could also: Reporting SD or a 95% confidence interval alongside or instead of SEM — SD directly reflects sample spread and CIs convey the precision of an estimated effect, which some readers find more informative than SEM, particularly for smaller group sizes typical of animal studies.
Software: Ingenuity Pathway Analysis (IPA) · GSEA (Gene Set Enrichment Analysis) · Nanostring nCounter analysis

What was reproduced

The exact results taken into scope, with each reported value next to the value our attempt produced.

C1_trem2_blocks_mgnd_sod1
Reported
Paper claims Trem2 loss prevents acquisition of the MGnD (disease-associated) signature in SOD1 mice; MGnD marker genes fail to upregulate in Trem2-KO vs Trem2-Het/WT.
Reproduced
Independent reanalysis of GSE102562 (RNA-seq processed matrix, TMM-normalized FPKM-like values as deposited): mean MGnD marker score (log2 mean(22 MGnD markers) - log2 mean(12 homeostatic markers)) = -2.777 (Trem2-KO, n=12) vs -1.035 (Trem2-Het, n=8). Welch t-tests on individual MGnD markers: Trem2 log2FC=-2.96 p=1.04e-07 FDR=3.5e-06; Lpl log2FC=-2.33 FDR=4.9e-05; Csf1, Axl, Ccl4, Itgax, Clec7a, Apoe, Cd63 all significantly down in KO (FDR<0.002). Homeostatic marker Tmem119 slightly up in KO (log2FC=+0.44, FDR=9.2e-04), consistent with retained/enhanced homeostatic identity.
within tolerance
C2_clec7a_pos_is_mgnd
Reported
Paper defines Clec7a positivity as a MGnD surface marker used to FACS-sort MGnD vs homeostatic microglia from APP-PS1 brains.
Reproduced
GSE102563 (RNA-seq, WT n=6, Clec7a-neg n=6, Clec7a-pos n=6): Clec7a-pos vs Clec7a-neg shows highly significant upregulation of Cd68 (log2FC=+1.65, FDR=7.1e-06), Lpl (+2.72, FDR=7.1e-06), Cd9 (+1.26), Ctsb (+1.90), Clec7a itself (+3.00, FDR=7.6e-06), Apoe (+3.31, FDR=9.2e-06), Axl (+2.52), Trem2 (+1.74, FDR=3.5e-05). NanoString GSE101687 group scores (mean MGnD score): Clec7a-negative=-2.351 (n=6) < Clec7a-intermediate=-1.678 (n=5) and Clec7a-positive=-1.954 (n=6) — negative clearly lowest, positive/intermediate both elevated but not perfectly ordered relative to each other.
within tolerance
C3_apoe_ko_impairs_phagocytosis_switch
Reported
Paper claims Apoe-KO microglia fail to fully transition to the MGnD state after phagocytosing apoptotic neurons, compared to WT.
Reproduced
GSE102564 (RNA-seq, WT/Apoe-KO x Phagocytic(P)/NonPhagocytic(NP)): WT_NP=-4.045 (n=10) -> WT_P=-1.765 (n=10), a large MGnD-score increase (+2.28) upon phagocytosis. KO_NP=-4.327 (n=10) -> KO_P=-3.003 (n=9), a much smaller increase (+1.32). WT_P (-1.765) is substantially higher than KO_P (-3.003), confirming attenuated MGnD induction in Apoe-KO. Genotype sanity check: Apoe itself log2FC=+6.78 (WT_NP vs KO_NP, FDR=5.6e-04), confirming correct genotype labeling of columns.
within tolerance
C4_phagocytosis_induces_mgnd_wt
Reported
Paper's central mechanistic claim: phagocytic uptake of apoptotic neurons drives the M0-to-MGnD transcriptional switch via TREM2-APOE signaling.
Reproduced
RNA-seq (GSE102564) strongly confirms this in WT: NonPhagocytic=-4.045 vs Phagocytic=-1.765 (large, significant increase; homeostatic markers Tgfbr1/Selplg/Tmem119/P2ry12/Sall1 all significantly down, FDR<2e-5, in Phagocytic vs NonPhagocytic). HOWEVER, NanoString reanalysis of the independently deposited RCC files shows the OPPOSITE direction in both relevant series: GSE101686 WT_Phagocytic_deadneurons=-4.455 (n=4) vs WT_NonPhagocytic_deadneurons=-1.384 (n=4); GSE101688 WT_Phagocytic=-3.470 (n=3) vs WT_NonPhagocytic=-0.574 (n=3). Both NanoString series consistently rank Phagocytic BELOW NonPhagocytic, contradicting both the paper's claim and the RNA-seq result from the companion series.
did not match
C5_eae_severity_dose_response
Reported
Paper claims microglial MGnD signature increases with EAE clinical disease score.
Reproduced
GSE101688 NanoString group means by EAE score: score0=-3.318 (n=3) < score1=-2.524 (n=4) < score2=-2.230 (n=6) < score3=-1.958 (n=6) < score4=-2.173 (n=1). Clean, essentially monotonic increase from score0 through score3; score4 (n=1, single sample) breaks strict monotonicity but is within the range of score2/score3.
within tolerance
C6_sod1_als_and_app_ps1_progression
Reported
Paper claims progressive MGnD signature acquisition with ALS disease score and with APP-PS1 age/plaque burden.
Reproduced
GSE101686 NanoString: SOD1 disease-score samples are mostly n=1 per score per animal, precluding meaningful group statistics (e.g. SOD1_01_score1=-0.823, SOD1_01_score2=-0.782, SOD1_01_score3=-0.702, SOD1_02_score1=-0.521, SOD1_02_score2=-1.550, SOD1_02_score3=-0.854 -- noisy, not cleanly monotonic within single animals). APP-PS1 age series: 2months=-4.869 (n=3, lowest, as expected) but 7months=-4.083, 10months=-3.308, 17months=-3.246 -- directionally increasing overall from 2 to 17 months but not strictly monotonic between 7/10/17.
partial

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

🤖 AI curator · claude (ai-curator room) · v1.0 L1 67/100

An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.

🟢1. Data identity
🟡2. Endpoint comparability
🟡3. Location of the main deviation
🟡4. Cause of the deviation
🟡5. Derivability / plausibility
🟡6. Severity of the deviation
🟡7. Core claim
🟡8. Severity of the miss (overall human judgment)
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Total score +10

Data identity is excellent — all six GEO series were pulled 1:1 from the authors' deposits (63 + 56 RCC files, three processed RNA-seq matrices), so nothing here is blocked by availability. The reproduction confirms the TREM2-APOE/MGnD backbone convincingly on RNA-seq: Trem2-KO blocks the switch (score -2.777 vs -1.035; Trem2 log2FC=-2.96, FDR=3.5e-06), Clec7a-pos = MGnD (Apoe +3.31, Lpl +2.72, Clec7a +3.00, all FDR<1e-05), Apoe-KO attenuates the phagocytosis-induced switch (+1.32 vs +2.28), and EAE severity tracks MGnD monotonically from score0 to score3. The one real problem is C4: both NanoString series (GSE101686, GSE101688) put phagocytic microglia below non-phagocytic, reversing both the paper and the companion RNA-seq result — but the reproduction's own notes identify its simplified 6-housekeeping-gene normalization, n=3-4 groups, and surrogate marker panel as the leading explanations, and no paper-reported value was ever placed against our numbers. This sits on our methodology side, not on the authors': the deviation is real and honestly flagged, yet not sufficient to call the central conclusion unconfirmed or the reported values non-derivable — hence yellow across the board rather than red.

🤝
Reproduced automatically — and fairly

Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.

Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.

🚩 Report an error in this record

Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.

Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.