Fraud detection report: 'Male-derived PBP4 is essential for sperm competition by mediating sperm motility in moths' (PNAS, DOI: 10.1073/pnas.2510155122)

Academic fraud report · Geng Detector

Summary

Verdict: Highly suspicious (orange). Three major textual and quantitative inconsistencies were identified, not relying on image forensics. First, the definition of sperm motility grades is inverted between the Results and Materials and Methods sections: Results defines Grade I as <50% flagellar oscillation and Grade II as >50%, whereas Methods states the exact opposite. This 180-degree swap would reverse the paper's core claim that PBP4 loss reduces Grade II sperm. Second, RNA-seq reports a 167-fold upregulation of Orco in PBP4 knockouts, while qPCR validation reports an 8,643-fold change—roughly 50× larger—an implausible gap within a single compensatory response. Third, Figure 5's legend defines 'ns' (no significant difference) with the condition P < 0.05, a logical and statistical contradiction. Together these suggest sloppy construction, possible reuse of text from unrelated manuscripts, or selective data handling. No image manipulation analysis was included in this report. Final determination requires institutional investigation and raw-data inspection.

Verdict

Highly suspicious. The paper contains internally contradictory definitions and biologically implausible quantitative gaps, in addition to a clearly erroneous statistical legend.

Key findings

Inverted definition of sperm motility grades between Results and Materials and Methods (180° swap of Grade I vs Grade II criteria).
Massive quantitative discrepancy: RNA-seq fold-change of 167× for Orco versus qPCR fold-change of 8,643× (~50× gap) within the same comparison.
Figure 5 legend misuses 'ns' by pairing 'no significant difference' with the threshold P < 0.05, a logical contradiction.
Issues are textual/quantitative; no image manipulation evidence was assessed.

Evidence highlights

Sperm grade inversion: Results states "Grade I motile sperm were characterized by less than half of the flagellar length displaying oscillation, while grade II motile sperm oscillated with more than half of the flagellum." Materials and Methods (Sperm Motility Analysis) states "Grade I motile sperm was designated as more than half of the flagellum oscillating, and grade II motile sperm was less than half of the flagella oscillating." Definitions for Grade I and Grade II are fully reversed.
Omics discrepancy: RNA-seq reports 167× upregulation of Orco in PBP4 knockout; qPCR validation reports 8,643× upregulation of Orco. The ratio qPCR/RNA-seq ≈ 51.7×.
Statistical legend error: Figure 5 caption states "'ns' indicates no significant difference with the CK (Student's t test, P < 0.05)." 'ns' conventionally requires P ≥ 0.05.
DOI: 10.1073/pnas.2510155122 (PNAS, Vol. 122, No. 44, e2510155122, 2025).

Notes

Findings rely on textual analysis; no Western blot, gel, microscopy, or statistical output forensics were performed here.
The inverted grade definition, if applied as written in the Methods section, would directly invalidate or reverse the central conclusion regarding Grade II sperm reduction in mutants.
The RNA-seq vs qPCR discrepancy could in principle reflect biological or technical differences, but a ~50× gap is unusually large and warrants raw-data inspection (raw counts, Cq values, amplification curves).
The 'ns' legend error may be a copy-paste artifact from a different manuscript, but raises concerns about author oversight of figure preparation.
Recommend requesting raw sperm motility scoring records, qPCR raw Cq values and standard curves, RNA-seq count matrices, and ethics/integrity statements.
Report is AI-assisted; final determination requires institutional investigation. Possible false positives exist.