Abstract
We incorrectly named the products of two genes, gamma actin and betaAPP, as FMRP ligands because of a misreading of Table 1 from Denman (2003). BetaAPP is incorrectly referred to as an FMRP ligand in Tables 2 and S9. This does not change the use of FMRP binding as a way to support the target predictions, as the statistics remain virtually the same; the enrichment of FMRP targets within the predicted microRNA target set changes from 5.31 to 5.26 as a result of our misclassification. In addition, it does not change the fact that betaAPP is a predicted microRNA target. Our Web site at http://www.microrna.org/ contains the updated information with changes flagged.
Introduction
In the field of genomics, identifying microRNA (miRNA) targets is essential for understanding post-transcriptional regulation. A foundational 2004 study published in PLoS Biology established a computational framework for these predictions. To validate the biological accuracy of the model, the authors utilized the Fragile X Mental Retardation Protein (FMRP) as a benchmark. This article details a formal correction regarding the classification of specific genes used in that validation process.
The Misclassification Error
The validation step of the original research relied on a list of FMRP ligands sourced from existing literature, specifically Table 1 of Denman (2003). Upon re-examination, it was discovered that two genes—gamma actin and betaAPP—were erroneously categorized as FMRP ligands due to a misinterpretation of the reference data.
As a result:
- BetaAPP was incorrectly listed as an FMRP ligand in Tables 2 and S9.
- Gamma actin was incorrectly listed as an FMRP ligand in Table S9.
Impact on Statistical Validation
The core of the study's validation was an enrichment analysis. The authors sought to determine if their predicted miRNA targets were significantly enriched with known FMRP ligands. A high enrichment score serves as evidence that the prediction algorithm captures genuine biological interactions.
Following the reclassification of gamma actin and betaAPP, the enrichment statistics were recalculated. The enrichment of FMRP targets within the predicted miRNA target set shifted from 5.31 to 5.26. This negligible change demonstrates the robustness of the original findings; the statistical signal is a broad biological trend and is not reliant on the inclusion of these two specific genes.
Conclusion
The authors emphasize that while betaAPP is no longer classified as an FMRP ligand in this context, it remains a predicted microRNA target. The fundamental conclusion of the original paper—that the computational method successfully identifies biologically relevant targets—remains valid. Updated information and flagged changes are available at the study's official repository, microrna.org.
Summary updated