Integrity Review Report: “RNA干扰HCCR-2基因表达对食管癌EC9706细胞凋亡及周期的影响”

Academic fraud report · Geng Detector

Summary

This review rates the paper highly suspicious, but the evidence is insufficient to establish fabrication conclusively without the authors’ raw data and images. The principal quantitative finding concerns the reported analyses of variance in Tables 1–3. For Table 1, the example values for HCCR-2 mRNA expression are 0.19 ± 0.07, 0.43 ± 0.22, and 0.45 ± 0.21, with n=3 per group. Recalculation from the reported means, standard deviations, and sample sizes gives an approximate F value of 1.93 and P value of 0.22, whereas the paper reports F=68.473 and P<0.01. Similar internal inconsistencies are reported throughout Tables 1–3, including P21 and P27. A second anomaly is the G418 concentration of 500 mg/ml for selection and 200 mg/ml for maintenance, which exceeds conventional mammalian-cell working concentrations of approximately 200–1000 μg/ml by an asserted 500–1000-fold; the unit is therefore a serious concern, although the report provides no independent validation of the stated standard range. Image duplication could not be assessed because no original Western blot or flow-cytometry images were supplied. The findings warrant an institutional and journal investigation, but are not by themselves a definitive finding of research misconduct.

Verdict

Highly suspicious. The reported statistical results in Tables 1–3 appear incompatible with the displayed means, standard deviations, and sample sizes. The stated G418 concentrations also require urgent clarification. No conclusion about image reuse or manipulation can be reached from the supplied text alone.

Key findings

The paper reports one-way analyses of variance with very large F values and uniformly significant results, including F=68.473 and P<0.01 for HCCR-2 mRNA expression in Table 1.
For the example data in Table 1—0.19 ± 0.07, 0.43 ± 0.22, and 0.45 ± 0.21, with n=3—reverse calculation yields approximately MS_between=0.0628, MS_error=0.0325, F=1.93, and P=0.22.
The report alleges that the same discrepancy occurs in P21, P27, and all data in Tables 2 and 3.
The methods section reportedly uses G418 at 500 mg/ml for selection and 200 mg/ml for maintenance. These units are stated to be inconsistent with conventional mammalian-cell concentrations, although the cited reference range was not independently verified here.
No Western blot bands, flow-cytometry plots, or other image files were available for pixel-level duplication or editing analysis.

Evidence highlights

Table 1, HCCR-2 mRNA expression:
Experimental group: 0.19 ± 0.07
Empty-vector group: 0.43 ± 0.22
Control group: 0.45 ± 0.21
n=3
Reported: F=68.473, P<0.01
Recalculated approximately: F=1.93, P=0.22
Methods, cell culture and transfection:
Reported G418 selection concentration: 500 mg/ml
Reported G418 maintenance concentration: 200 mg/ml
Image analysis status: not assessable because original visual evidence was not provided.

Notes

The report assesses the supplied materials and should not be treated as a final misconduct determination. The authors should be asked to provide raw numerical data, analysis files, laboratory records, and original unedited images. The journal and relevant institution should independently recalculate every statistical table, verify the G418 units and experimental protocol, and inspect the underlying images. The confidence is high for the reported mathematical inconsistency in the Table 1 example, moderate for the broader pattern across the other tables, and low for image-related conclusions because no images were available.