Investigation Report: Neuronal Sensitization Drives Bone Regeneration via Energy Metabolism and BMP2-ErbB1 Signaling (DOI: 10.1002/adfm.76607)

Academic fraud report · Geng Detector

Summary

This report evaluates the paper "Neuronal Sensitization Drives Bone Regeneration via Energy Metabolism and BMP2-ErbB1 Signaling" published in Advanced Functional Materials (DOI: 10.1002/adfm.76607) by Dai et al. The overall verdict is highly suspicious. The most serious finding concerns an instrumentation error: the methods state that ion concentrations were measured using an ICP-OES (PerkinElmer PinAAcle 900T, Germany); however, the PinAAcle 900T is in reality a flame/graphite-furnace atomic absorption spectrometer (AAS), not an ICP-OES (which PerkinElmer markets as Optima or Avio). This indicates possible fabrication of methodology. Additional concerns include a likely ethics-approval mismatch (approval no. SYT2025040 prefixed in a manner suggesting Shenzhen Institutes of Advanced Technology rather than South China University of Technology), omission of BC and Mg-BC control groups from the formulation table (Table 2) despite their repeated use in figures, an inconsistency between n=4 biological replicates stated for RNA-seq versus n=3 stated for all other cell experiments, and a weak citation (reference 16) concerning erectile dysfunction used to support claims about peripheral nerve development. Confidence is moderate-to-high on text-based findings, but image-based forensic analysis was not possible without the original figures.

Verdict

🟠 Highly suspicious. Multiple text- and method-level irregularities, including a definitive instrument-principle mismatch, raise credible concerns about research integrity. Image-level forensics were not performed.

Key findings

Instrument-principle mismatch (🔴 confirmed): Methods cite "ICP-OES, PerkinElmer PinAAcle 900T" for ion detection. The PinAAcle 900T is an atomic absorption spectrometer (AAS), not an ICP-OES.
Possible ethics-approval misattribution (🟠 insufficient evidence): Approval no. SYT2025040 is attributed to South China University of Technology, but the "SYT" prefix is consistent with Shenzhen Institutes of Advanced Technology (SIAT) internal numbering.
Omitted control formulations (🟠 confirmed): Table 2 lists only BMP2@P, BMP2-BC, and Mg-BMP2@P-BC, while BC and Mg-BC controls appear throughout Figures 2–3 without their recipes disclosed.
Replicate-number inconsistency (🟡 confirmed): RNA-seq uses n=4 biological replicates, while all cell-experiment figure legends state "independently repeated at least three times (n=3)."
Tangential / misfit citation (🟡 confirmed): Reference [16] (Kwon et al., on BMP2 in erectile dysfunction in diabetic mice) is cited to support BMP2's role in peripheral nerve development and regeneration.
Image forensics not performed (ℹ️): No pixel-level duplication or splicing analysis was conducted due to lack of figure files.

Evidence highlights

Direct quotation of method (p. 15): "…were detected using an inductively coupled plasma optical emission spectrometer (ICP-OES, PerkinElmer PinAAcle 900T, Germany)."
Ethics statement (p. 16): "approved by the Laboratory Animal Research Centre of South China University of Technology (approval no: SYT2025040)."
Replicate discrepancy (p. 16, §5.13): "…four independent biological replicates (n=4) for transcriptome RNA-seq analysis."
Figure legend standard (Figures 2–4): "All the experiments were independently repeated at least three times (n=3); data are presented as mean±SD."
Citation [16] content: "BMP2 Restores Erectile Dysfunction through Neurovascular Regeneration… in Diabetic Mice," Andrology.

Notes

The PinAAcle 900T instrument misidentification is the most concrete and verifiable finding; this kind of error is unusual for researchers who actually performed the measurement.
The ethics-approval concern is plausible but cannot be confirmed without cross-checking the issuing institution's internal numbering system; reporting institutions should be queried.
The BC / Mg-BC omission is unusual but could in principle reflect an oversight rather than intentional concealment; reproducibility is nevertheless compromised.
The n=3 vs. n=4 discrepancy may reflect separate experimental batches rather than fraud, but the absence of explanation in the methods is itself a reporting deficiency.
Reference [16]'s topical mismatch with the supporting claim is a citation-quality issue; whether it constitutes misconduct depends on the full text of [16] and surrounding context.
Recommended follow-ups: request raw RNA-seq fastq files and sequence-submission records; request original ICP/OES instrument logs; verify the issuing body of ethics approval SYT2025040 with both South China University of Technology and SIAT; raise concerns on PubPeer and notify the editorial office.
DOI of the article under review: 10.1002/adfm.76607.
Limitations: image-based checks (duplication, splicing, background-noise analysis) were not possible from the supplied text-only PDF.