Researchers have retracted a 2015 Nature paper about the molecular underpinnings of immune function after discovering they could not replicate key parts of the results.
The first author, Wendy Huang — who started working as an assistant professor at the University of California, San Diego, only months after the paper appeared — did not sign the retraction letter, published last week. The research was conducted while Huang was working as a postdoctoral fellow at New York University, home of last author Dan Littman (also an investigator at the Howard Hughes Medical Institute).
What happened appears to be a case of “he said, she said:” Littman asked to retract the paper after his lab couldn’t reproduce it, and Huang insists the data remain correct, saying the process had been “unfair and done without due process:”
My attorney has sent letters to raise concerns…
Huang declined to provide more information regarding her attorney’s actions.
In follow-up experiments to this Article, we have been unable to replicate key aspects of the original results. Most importantly, an RNA-dependent physical association of RORγt and DDX5 cannot be reproduced and is not substantiated upon further analysis of the original data. The authors therefore wish to retract the Article. We deeply regret this error and apologize to our scientific colleagues.
The 2015 paper has been cited 63 times, according to Clarivate Analytics’ Web of Science.
Huang’s research focuses on the role of the immune system in mouse models of colitis and multiple sclerosis. In 2016, Littman and Huang abandoned a patent covering methods to identify agents that alleviate some of the immune issues associated with these and other diseases, which incorporates the results of the now-retracted paper:
See also Huang et al. (2015, Nature 528:517-522), the entire content of which is incorporated herein by reference.
A collaboration gone wrong
According to Littman, the researchers tried to replicate the data when they tried to pursue the biochemistry of the reported interactions between the various molecules, and also construct the complex’s three-dimensional structure:
A new postdoctoral fellow spent a considerable amount of time trying to co-precipitate the complex but was unable to do so. What we previously interpreted as bands corresponding to the relevant proteins appear to be artifactual bands (likely immunoglobulin heavy chain). There is no reason to suspect foul play, but we think that there were artifacts introduced in the co-immunoprecipitations and there was also misinterpretation of some of the raw data for several of the key experiments.
Littman said Huang declined an invitation to come back to the lab to help reproduce the results, but provided all of her old data. Still, it wasn’t enough to convince Littman the results were valid.
Huang, however, said she’s been able to reproduce the data:
After arriving at UCSD, two postdoctoral fellows in my lab independently confirmed the essential role of DDX5 in cultured Th17 cells from our own mouse colony, confirmed its associated lncRNAs previous reported as well as identified new partner of DDX5, confirmed its role in RNA Pol2 recruitment genomewide, and assessed its contribution to chromatin accessibility…In Aug 2017, my ex-postdoc adviser emailed to state that they “had a great deal of difficulty in moving forward with the DDX5/Rmrp project”. I offered to help troubleshoot their experiments remotely. I inquired on the specifics of their challenges, including how repeat experiments were performed, their use of key reagents and protocol.
She said Littman never responded to her inquiries.
In January 2018, [Littman] requested that I sign a letter to retract our paper, stating that the published result “is not substantiated upon further analysis of the original data”. However, my reanalysis of available raw data together with other independent scientists have not identified miscalculations or experimental errors that warrant retraction suggested by [Committee on Publication Ethics] guidelines. On March 11, 2018, I wrote to the journal editor, providing results of repeat experiments from my current lab and others confirming numerous published findings, and again requested that the evidence supporting the irreproducibility claims be shared with myself and all authors. On June 1st 2018, the editor declined my request of evidence of non-reproducibility and also declined to publish my statement of dissent.
Huang sent us a copy of the form she submitted to Nature showing she disagreed with the retraction, which included this statement:
In an analytical replication study, the original data of the published work were re-analyzed by myself and multiple independent scientists and found that they do support the published findings. To my knowledge, numerous direct replication studies by others in multiple laboratories have confirmed that DDX5 and its associated lncRNA Rmrp modulate Th17 cell effector functions, just as the published title stated, and specific mutations of Rmrp altered its interaction with proteins and localization on chromatin DNA. Lastly, the communicating author has not provided me with any evidence supporting the claim of non-reproducibility.
Huang was co-first author on a 2015 paper in Cell that was corrected the following year over “three errors in measurements.” That Cell paper has been cited 112 times, making it a “highly cited paper,” meaning it was ranked in the top 1 percent of all papers in its field for the year it was published.
Littman is a scientific co-founder of Vedanta Biosciences, which designs medicines that act on the immune system to treat disease. In 2016, he won the Vilcek Prize in Biomedical Science, worth $100,000, offered to scientists who have immigrated to the U.S. He is a member of the board of directors at Pfizer.
Like Retraction Watch? You can make a tax-deductible contribution to support our growth, follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up for an email every time there’s a new post (look for the “follow” button at the lower right part of your screen), or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at email@example.com.