Data Fraud in Cancer Research at Harvard: Can ResearchHub mitigate this in the future?

Harvard and its affiliated Dana Farber Cancer Institute has retracted 6 papers from top journals after the whistleblower Sholto David found clear instances of data fabrication. They have gone further to amend 31 other published papers. Here, I provide an overview of the alleged fraud, and ask the ResearchHub community if decentralized science can help mitigate such fraud in the future. The highlighted figures with the fraud are from annotations made in this article by Leonid Schneider.

These examples here are simply the image fabrication ones, basic photoshop. They are still investigating the actual data and plots, which are harder to catch. What’s pretty insane is that this happened almost 20 yers ago, and these papers have subsequently accumulated countless citations by virtue of the author’s reputation and the journal’s reputation. This means that for almost 20 years, other well meaning scientists have spent millions, maybe even tens of millions of tax payer dollars, to use these results as ground truth and presumably go on a wild goose chase. This is seriously damaging to Harvard’s reputation, but also seriously damaging to the larger biomedical research community.

It raises questions about how we, as the community of scientists, can mitigate this in the future. Could Decentralized Science solve this problem, or would this problem persist regardless? If DeSci is the answer, can ResearchHub play a role?

Plasma cell differentiation and the unfolded protein response intersect at the transcription factor XBP-1 (Nature Immunology 2003)

DOI: https://doi.org/10.1038/ni907

XBP-1 specifically promotes IgM synthesis and secretion, but is dispensable for degradation of glycoproteins in primary B cells (Journal of Experimental Medicine 2005)

DOI: https://doi.org/10.1084/jem.20050575

Proapoptotic BAX and BAK modulate the unfolded protein response by a direct interaction with IRE1alpha (Science 2006)

DOI: 10.1126/science.1123480

Cancer-Associated PP2A Aα Subunits Induce Functional Haploinsufficiency and Tumorigenicity (Cancer Research 2005)

DOI: 10.1158/0008-5472.can-05-1103

A Lentiviral RNAi Library for Human and Mouse Genes Applied to an Arrayed Viral High-Content Screen (Cell 2006)

DOI: 10.1016/j.cell.2006.01.040

Identification of specific PP2A complexes involved in human cell transformation (Cancer Cell 2004)

DOI: 10.1016/s1535-6108(04)00026-1

Identification of specific PP2A complexes involved in human cell transformation (Nature 2003)

DOI: 10.1038/nature12147

Preprint

Preprint

Data Fraud in Cancer Research at Harvard: Can ResearchHub mitigate this in the future?