Geng Academic Fraud Report: "Extrachromosomal Circular DNA: A New Target in Cancer" (Frontiers in Oncology, 2022)

Academic fraud report · Geng Detector

Summary

Verdict: CLEAN (✅ 清白). This Frontiers in Oncology review article (DOI: 10.3389/fonc.2022.814504), authored by Wu, Liu, Zhou, Xiong et al. and received 17 March 2022 (published 14 April 2022), is a narrative review of extrachromosomal circular DNA (eccDNA) in cancer. All four Figures (eccDNA formation models, detection methods, expression regulation, cancer evolution) are author-drawn conceptual schematic diagrams; the paper contains no original experimental data (no Western blots, gels, flow cytometry, qPCR charts, bar graphs, survival curves, boxplots, or raw numerical tables). Consequently, most dimensions of the Geng six-form fraud detection framework are inapplicable to this article. The reference list contains 168 entries spanning 1965 (Cox et al., Lancet) to 2021 (Wang et al., Nature), author contributions and funding declarations (NSFC 81672683, Hunan NSF 2018SK21211, 111 Project 111-2-12) are complete, and no conflicts of interest are declared. All programmatic forensic signals (copy-move, noise variance, PRNU) that surfaced lacked any positive corroboration in the source text and were attributable to benign explanations such as shared export pipelines and journal recompression. Confidence in the clean verdict is moderate-to-high, limited by the inherent constraints of automated forensic tools on schematic illustrations rather than primary data images.

Verdict

✅ CLEAN — No substantiated findings of academic misconduct.

Key findings

The article is a Review (review article), not an original research paper; Figures 1–4 are author-drawn conceptual schematics, not primary experimental images.
The paper contains no Western blots, gel electrophoresis images, flow cytometry plots, qPCR data, bar graphs, survival curves, boxplots, or continuous numerical data tables — therefore most Geng six-form fraud-detection dimensions lack an applicable premise.
The reference list contains 168 entries (numbered 1–168), spanning Cox et al., Lancet (1965) through Wang et al., Nature (2021) and Morton et al., Cell (2019), with appropriate thematic coverage and chronological distribution.
Author contributions and funding declarations (National Natural Science Foundation of China 81672683; Natural Science Foundation of Hunan Province 2018SK21211; 111 Project 111-2-12) are clearly stated; no commercial conflict of interest is declared.
Manuscript timeline (submitted 2021-11-13; accepted 2022-03-17; published 2022-04-14) is reasonable and shows no unusual compression.

Evidence highlights

Figure captions (verbatim from source):
Figure 1 — "Four main models of eccDNA formation" (conceptual schematic).
Figure 2 — "Detection methods for eccDNA" (classification flowchart).
Figure 3 — "Regulation of eccDNA expression" (comparative schematic).
Figure 4 — "EccDNA accelerates cancer evolution" (conceptual schematic).
Reference count: 168 entries.
Funding identifiers preserved: 81672683; 2018SK21211; 111-2-12.
DOI: 10.3389/fonc.2022.814504.

Notes

Forensic signals (copy-move, noise variance, PRNU) that were generated automatically lacked positive corroboration in the source text. Per Geng review principle 8d, "unable to rule out" does not equate to a substantiated finding; nor does the absence of positive textual support for benign explanations (shared export pipeline, journal recompression) constitute exonerating evidence. In this review-article context, such signals are treated as inapplicable rather than as evidence of misconduct.
All originally flagged findings that could not be substantiated by data points present in the source text have been removed; remaining observations are informational only and serve to corroborate compliance (reference coverage, authorship disclosure, funding transparency).
This report is AI-assisted and intended for academic discussion only; final determination of misconduct requires investigation by qualified institutions.