Geng Integrity Report: Neuronal Sensitization Drives Bone Regeneration via Energy Metabolism and BMP2‐ErbB1 Signaling (Advanced Functional Materials, DOI: 10.1002/adfm.76607)

Academic fraud report · Geng Detector

Summary

This report evaluates the 2026 Advanced Functional Materials paper 'Neuronal Sensitization Drives Bone Regeneration via Energy Metabolism and BMP2‐ErbB1 Signaling' (Qiyuan Dai et al.; DOI: 10.1002/adfm.76607). The overall verdict is highly suspicious (🔴/🟠). Three textual and tabular concerns were verified. (1) In Table 1, the ZDOCK score for ErbB1 is reported as 13553.6, an order of magnitude higher than the other receptors (range ~1300–1900), which is physically implausible for a single docking batch and strongly suggests fabrication or a copy-paste format error. (2) The binding energy for ErbB1 and ErbB2 is identical (−17.7 kcal/mol), an unlikely coincidence for two distinct receptors. (3) Figure 3's legend misclassifies Smad1—an intracellular BMP-pathway transcription factor—as an 'osteogenic neurotransmitter' alongside CGRP, BDNF, and NGF, indicating careless framing. Image-level analyses could not be performed because only extracted text was available. Confidence in findings 1–3 is high; the molecular docking anomaly should be prioritized for editorial investigation. Pixel-level figure integrity remains unverified.

Verdict

Highly suspicious (🔴/🟠). Three textual/tabular integrity issues were verified against the source PDF. Two concern Table 1 (molecular docking data) and one concerns a basic biological mislabeling in Figure 3. Pixel-level image forensics were not performed because only text-extracted content was available.

Key findings

ZDOCK score outlier for ErbB1 (finding 1, 🔴): ErbB1's ZDOCK score (13553.6) is roughly an order of magnitude larger than all other receptors in Table 1, which cluster between ~1300 and ~2000. Such an intra-batch deviation exceeds expected physical/statistical ranges of the ZDOCK scoring function and is highly indicative of data fabrication or a copy-paste error that escaped proofreading.
Identical binding energies for ErbB1 and ErbB2 (finding 2, 🟠): Table 1 lists the binding energy of both ErbB1 and ErbB2 as −17.7 kcal/mol. Two distinct receptors yielding binding energies identical to one decimal place is statistically implausible and points to fabricated/duplicated numbers.
Conceptual mislabeling of Smad1 as a neurotransmitter (finding 3, 🟡): Figure 3's legend groups Smad1 with CGRP, BDNF, and NGF under 'osteogenic neurotransmitters.' Smad1 is a canonical intracellular BMP-pathway R-Smad transcription factor, not a secreted neurotransmitter, reflecting clear terminological/conceptual carelessness.

Evidence highlights

Table 1 (Page 9) – ZDOCK scores as printed: BMPR1A = 1528.46; BMPR1B = 1967.40; BMPR2 = 1601.76; ErbB1 = 13553.6; ErbB2 = 1937.10; ErbB3 = 1778.69; ErbB4 = 1879.33.
Table 1 (Page 9) – Binding energies (kcal/mol) as printed: −13.5 (BMPR1A), −12.6 (BMPR1B), −10.9 (BMPR2), −17.7 (ErbB1), −17.7 (ErbB2), −14.7 (ErbB3), −18.2 (ErbB4).
Figure 3 legend: States 'Expression of osteogenic neurotransmitters (CGRP, BDNF, NGF, Smad1) in SH-SY5Y cells,' improperly classifying Smad1 as a neurotransmitter.
DOI: 10.1002/adfm.76607 (Advanced Functional Materials, 2026).

Notes

Image-based checks (Western blot splicing in Figure 6F; immunofluorescence in Figures 3A/C, 4, 9; histology in Figure 8) were not possible because only text-extracted content was supplied to the reviewer. Findings 1–3 are text/table-only and should be re-verified against the original published PDF and any source data files.
The ErbB1 ZDOCK outlier (13553.6 vs. ~1300–2000 for peers) is the single highest-priority anomaly and should be raised with the corresponding author and the journal's editorial office, along with a request for raw PDB files, docking logs, and the underlying scoring outputs.
All verdicts are preliminary and based solely on the materials provided; official institutional investigation is required for any formal determination of misconduct.