Correction: Corrigendum: Systematic pan-cancer analysis of tumour purity

Dvir Aran, Marina Sirota, Atul J. ButteNature Communications 2016 · DOI · OpenAlex

Introduction

This document serves as a formal corrigendum for the 2015 study "Systematic pan-cancer analysis of tumour purity," published in Nature Communications. The original research provided a comprehensive framework for assessing tumor purity—the proportion of malignant cells relative to stroma, immune cells, and other non-cancerous elements—which is a critical confounding variable in genomic analysis.

The Correction

A nomenclature error was identified in the Results section of the original Article regarding the histological validation data sourced from The Cancer Genome Atlas (TCGA) repository. The original text incorrectly referred to these validation slides as being produced via immunohistochemistry (IHC).

This corrigendum clarifies that the data actually utilized haematoxylin and eosin (H&E) staining. Consequently, all instances of "immunohistochemistry" or "IHC" in the original paper should be read as "haematoxylin and eosin staining" or "H&E staining," respectively.

Significance of the Distinction

While this is a textual correction, the distinction between H&E and IHC is significant for researchers in oncology and pathology.

  • H&E Staining: This is the "gold standard" for general tissue morphology. Haematoxylin stains cell nuclei blue/purple, while eosin stains the cytoplasm and extracellular matrix pink. It is the standard method used by pathologists to visually estimate the percentage of tumor cells in a bulk sample (tumor cellularity).
  • Immunohistochemistry (IHC): IHC uses specific antibodies to target and visualize the presence of particular proteins. While IHC can identify specific cell types (e.g., using CD8 to identify T-cells), it is not the standard "all-purpose" stain used for the broad morphological assessments typically found in the TCGA's primary diagnostic slides.

Impact on Reproducibility

For researchers attempting to replicate the study's validation steps or benchmark new computational purity tools (such as ESTIMATE or ABSOLUTE), knowing the exact staining method is vital. Computational purity estimates are often benchmarked against these visual H&E estimates. This correction ensures the technical accuracy of the "Results" section regarding how the TCGA repository data was generated and utilized for validation.

Conclusion

This update ensures that the technical terminology aligns with the actual laboratory procedures used to generate the validation data. The fundamental findings and computational models of the original 2015 study remain valid, but the description of the histological evidence is now accurately represented as H&E staining.

Summary updated

Abstract

Nature Communications 6 Article number: 8971 (2015); Published 4 December 2015; Updated 5 February 2016 In the Results section of this Article, haematoxylin and eosin staining data from the TCGA repository are incorrectly referred to as immunohistochemistry data. All references to ‘immunohistochemistry’ or ‘IHC’ should read ‘haematoxylin and eosin staining’ or ‘H&E staining’, respectively.
BibTeX
@inproceedings{aran2016correction,
  author = {Dvir Aran and Marina Sirota and Atul J. Butte and Atul Butte},
  title = {{Correction: Corrigendum: Systematic pan-cancer analysis of tumour purity}},
  booktitle = {Nature Communications},
  year = {2016},
  doi = {10.1038/ncomms10707},
}
Figure