University of Minnesota research is under review following a rare retraction.

Veterinary medicine professor Sagar Goyal published a retraction in December, withdrawing a research article from 2010. The University’s Academic Health Center is now evaluating a research misconduct complaint to find out what happened and why.

Retractions in scientific publications are rare. A 2012 study found about 4,450 scientific articles have been retracted since 1928. According to a 2008 study, 1.35 million academic articles were published in 2006 alone.

Of the retractions, only 20 percent were for research misconduct.

Goyal and eight other researchers previously published their identification of four new swine flu strains in birds in the journal Emerging Infectious Diseases. The retraction also withdrew the strains from GenBank, a national register.

The Minnesota Center of Excellence for Influenza Research and Surveillance at the University funded the research with money from the National Institutes of Health.

AHC spokesman Justin Paquette said an NIH researcher worked with the swine flu strains that Goyal and his colleagues had identified, and the researcher noticed that they were different from their description in the article.

Someone from the University filed an official research misconduct complaint, but Paquette said it’s just an allegation. The University is currently examining the complaint before deciding if a formal review of the research is necessary.

“The U of M obviously takes any such concerns seriously and wants to do all due diligence around the work in question,” Paquette said in a statement.

Goyal has worked at the University for 31 years. He has published more than 250 research articles and won the Civil Service Award for Outstanding Faculty in 1991.

James Collins, director of the Veterinary Diagnostic Laboratory, has worked with Goyal since he started at the University. To non-researchers, Collins said, retractions and complaint inquiries may seem “alarming,” but they’re part of the self-corrective research process.

“This is [a] pretty standard kind of thing that people do in the scientific community when there’s intellectual discovery or debate,” he said.

Retraction action

Most retractions occur in medical research, according to a 2012 study.

The main reason for publishing retractions was because of “publishing misconduct,” which could be plagiarism or a duplicate publication with the same findings.

At the University, retractions are “infrequent,” according to an email from Associate Vice President for Research Frances Lawrenz.

“Retraction is an important part of the scientific process as it helps ensure the integrity of process itself,” she said in the email. “It is nearly impossible, however, to estimate how often articles are retracted.”

University General Counsel Mark Rotenberg said central administration doesn’t keep track of retractions because University faculty and staff publish hundreds of articles every month.

The Office of the General Counsel handles a “miniscule” number of inquiries for research misconduct, he said.

If an inquiry warrants a full investigation, Rotenberg said the University reports its findings to funding agencies, which can then conduct their own investigations. The University then determines what kind of discipline is necessary.

“It is vitally important for the University to maintain public confidence that our research is high quality, unbiased and objective,” Rotenberg said. “We take very seriously allegations that research has not been undertaken or published in a way that adheres to the highest standards of accuracy and objectivity.”

A unique case

After the NIH researcher discovered something was different about the swine flu strains, the researcher notified Mike Osterholm, director of the center that funded the research, indicating the strains might have been mislabeled or cross-contaminated.

Osterholm then asked an external expert to review the strains, who agreed mislabeling or cross-contamination was likely the problem, Paquette said.

Then the retraction was printed, removing the strains from GenBank.

Paquette said this order of events is unique for this case because a retraction is usually printed after going through the inquiry process and identifying research misconduct.

“[T]he University is conducting an assessment of related research to determine whether additional retractions may be necessary until we can assess what led to possible contamination, mislabeling or any other issue that led to the original retraction,” Paquette said in the statement.

Collins said he has faith the University will do the right thing, whatever it may be.

“I’m sure that the University will look into it and do whatever is in the best interest of the research community and society.”