Abstract
This article serves as a formal publisher correction for two significant papers published in Nature Methods: the analysis of the 2018 Data Science Bowl and the Human Protein Atlas Image Classification competition. The primary purpose of this amendment is to transition these works to an Open Access status under the Creative Commons Attribution 4.0 International License.
Introduction
In February 2020, Nature Methods (Volume 17) issued a formal correction notice regarding two major bioimaging competition papers. These papers summarized the outcomes of large-scale community efforts: the 2018 Data Science Bowl, focused on nucleus segmentation, and the Human Protein Atlas Image Classification challenge, focused on protein localization. While the original publications provided critical insights into automated biological imaging, their initial licensing restricted broad accessibility.
The Shift to Open Access
The core of this correction is the administrative transition of these papers from standard Springer Nature copyright to a Creative Commons Attribution 4.0 (CC BY 4.0) license. This change reflects the collaborative and open-source nature of the competitions themselves, which relied on the contributions of thousands of data scientists and researchers worldwide.
By adopting the CC BY 4.0 license, the publisher ensures that the collective knowledge generated by these competitions is freely available. This license permits users to:
- Share: Copy and redistribute the material in any medium or format.
- Adapt: Remix, transform, and build upon the material for any purpose, even commercially.
The only requirement is providing appropriate credit to the original authors and indicating if changes were made.
Significance for Bioimaging Research
The two papers affected by this correction represent milestones in the application of deep learning to biological imaging:
- 2018 Data Science Bowl: Addressed the challenge of segmenting nuclei across diverse imaging modalities.
- Human Protein Atlas Image Classification: Addressed the multi-label classification of protein patterns in human cells.
Making these summaries Open Access allows the global research community to build upon the winning architectures and methodologies without the barriers of a paywall, fostering further innovation in automated cell biology.
Conclusion and Reference Updates
This correction does not alter the scientific findings or data presented in the original studies. It serves as a legal update to ensure the papers meet the needs of the scientific community for open and reproducible research. The specific corrections can be accessed via the following DOIs:
- Nucleus segmentation correction: 10.1038/s41592-020-0733-z
- Human Protein Atlas correction: 10.1038/s41592-020-0734-y
Summary updated