Broad Institute of MIT and Harvard, 75 Ames St, Cambridge, MA 02142, USA.
Department of Pathology and Center for Cancer Research, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA.
Gigascience. 2021 Mar 12;10(3). doi: 10.1093/gigascience/giab015.
To enhance reproducibility in scientific research, more and more datasets are becoming publicly available so that researchers can perform secondary analyses to investigate questions the original scientists had not posited. This increases the return on investment for the NIH and other funding bodies. These datasets, however, are not perfect, and a better understanding of the assumptions that shaped them is required. The 2020 Junior Research Parasite Award recognized our work that showed that the signal-to-noise ratio in a particular dataset had not been investigated, leading to an erroneous conclusion in the original research. In this commentary, I share the process that led to the identification of the problem and hopefully provide useful lessons for other research parasites.
为了提高科学研究的可重复性，越来越多的数据集变得公开可用，以便研究人员可以进行二次分析，以研究原始科学家没有提出的问题。这增加了 NIH 和其他资助机构的投资回报。然而，这些数据集并不完美，需要更好地了解塑造它们的假设。2020 年青年研究寄生虫奖认可了我们的工作，该工作表明，特定数据集中的信噪比尚未得到研究，导致原始研究得出错误的结论。在这篇评论中，我分享了导致问题识别的过程，并希望为其他研究寄生虫提供有用的经验教训。
