The news isn’t getting any better for Carsten Carlberg. Earlier this month, Carlberg received a pink slip from the University of Luxembourg in the wake of a data fraud scandal involving a former student named Tatjana Degenhardt. As we wrote last November, journals retracted two of Carlberg’s papers after Degenhardt was found to have fabricated data. Although Degenhardt worked at the University of Kuopio (now the University of Eastern Finland), and Carlberg was not accused of wrongdoing, officials in Luxembourg decided that the mess threatened the public image of their fledgling institution and fired him.
We’d assumed that the University of Eastern Finland considered the matter finished, and that what happened in the Grand Duchy would stay in the Grand Duchy. But we were wrong.
The University of Eastern Finland appears to have opened an ethics investigation into Carlberg and is considering whether to fire him, too, according to a recent story in Savon Sanomat (the architecture of their site makes it very difficult to post to the actual story). Meanwhile, the school has voided Degenhardt’s doctoral dissertation, which she earned under Carlberg’s supervision and which landed her a post-doc in the MIT lab of Ernest Fraenkel. She left last summer, but when we first reported on this case, neither she nor Fraenkel would say whether her deception was the reason for her departure.
At that time, Degenhardt told us that
I am very sorry that it happened, but is also very difficult to explain.
Since neither of us speaks Finnish (or for that matter Swedish, which also is spoken there), we’re relying on translations from readers for this post.
New retraction, Interesting how they wrote it.
http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2567.2011.03491.x/abstract
Thanks Ari — we’ve now posted on that unrelated one, too: http://www.retractionwatch.com/2011/07/29/university-investigating-leading-german-derm-lab-that-retracted-immunology-paper/
This is a full translation of the article published in Savon Sanomat on July 22, 2011. There appear to be some misunderstandings of the reporter (such as the 50+50 % positions in Kuopio and Luxemburg, and the position of professor Carlberg as the head of Cardiac and Vascular and type II diabetes reseach unit), which I did not try correct. So, this is how the case was told in Savon Sanomat.
– RESULTS MANIPULATED
– Professor of University of Eastern Finland has been fired, doctoral degree withdrawn, and other PhD theses have been examined.
A rare fraud in PhD thesis has been revealed at the University of Eastern Finland. As its consequence professor Carsten Carlberg from the Department of Biosciences at the University of Eastern Finland has been fired from his position at the University of Luxemburg starting from the beginning of July. Also the University of Eastern Finland considers his dismissal. Carlberg has half position in both universities.
”The process is almost finished, although the holiday season slows it down. The chain of events has been followed in cooperation with the University of Luxemburg. We will hear Carlberg once again before decision”, tells the dean of the Faculty of Science and Forestry Timo Jääskeläinen from the University of Eastern Finland.
A special ethical commission has been constituted, led by the academic rector Kalervo Väänänen, to solve the case. The University of Eastern Finland has been working on the case since last November.
Carlberg himself has not been found guilty on the fraud, but a PhD student Tatjana Degenhardt who worked and studied under his supervision falsified the results of the laboratory experiment into pleasing ones for her PhD thesis. Degenhardt’s thesis was accepted in 2009 at the University of Kuopio. Now this thesis has been cancelled, and Degenhardt has admitted the fraud in writing. ”We cancelled the thesis just before midsummer and her doctoral degree was disqualified”, tells Jääskeläinen.
Carlberg as the supervisor of his research group should have monitored the work of his student. The responsibility question is clear, thus, the University of Luxemburg considered Carlberg had smeared the reputation of their University. During the hearing, other group members have been heard and approximately ten PhD theses have been examined. ”No problems were found in them. This appers to be a single case”, says Jääskeläinen.
The thesis which contained the manipulated results had five articles published in international scientific journals. The fraud was noticed last year by the international science community, which found out that the reported experimental results could not be obtained. For the University of Eastern Finland the fraud was revealed via Retraction Watch journal, which informs the observed drawbacks in published experiments. ”Two of the published articles had to be retracted due to the tampering of the experimental results”, tells Jääskeläinen.
Besides Degenhardt there were also other authors in the published articles. ”In the hearing Tatjana took the rap for fraud. She has done the experiments and manipulated the data. The hearing process has been strict, since for a young person it is big thing when a degree is taken away.” According to Jääskeläinen, the hard competition among the young scientists is behind the case. ”There has been huge pressure on getting results, and the moral has fallen down”.
Degenhardt, who no longer lives in Finland, says she is very sorry that it happened, but says it is also very difficult to explain. Carlberg himself has accused Degenhardt as a traitor student, whom he thought he knew well. ”A year ago I would have put my hand into fire for her” states Carlberg in Retraction Watch.
According to Jääskeläinen, the case is rare. ”During my 30-year career I have not experienced anything similar”. Neither does the university rector Perttu Vartiainen remember a case, where the thesis would have been withdrawn due to falsifying. ”I do not want to mark Germany in any way, but there just among the politicians there are a half-a-dozen theses which have been withdrawn due to plagiarisms. There are number of these cases in Germany, but this one is different”, says Vartiainen.
The PhD thesis process in Finland is strict. The cases of plagiarism are more easily discovered, however, the monitoring of single experimental results is extremely difficult. ”Fiddlings are revealed afterwards when the scientific community repeats the experiments”, notes Jääskeläinen.
The manipulation of the reseach results occurred at the research unit of Cardiac and vascular diseases and type II diabetes of the University of Kuopio, that was led by professor Carlberg. The thesis was on molecular biology with PPAR target genes and metabolic routes as a topic. Doctor Christopher M. Bunce from Birmingham University was the opponent. As custos was professor Carsten Carlberg.-
Why don’t you pay attention to two papers, which prof. Carsten Carlberg are involved with: In BBA 1813(5): 810-818, 2011 Fig. 3 E and BBA 1803(11): 1276-1286 Fig. 2B the graphs for two different genes (IL-10 and IL-.12) look surpirisingly similar.
Sepi
After enlarging one of the figures a little, I could achieve a perfect overlay in Photoshop. Did you notify BBA?
TK, you are right. Figures 3E 1813 BBA 2011 and 2B BBA 1803 2010 are perfectly matching. It seems that this escaped to the attention of the external commission in Luxembourg who examined BBA 1803 (2010) 1276-1286. The commission expressed concern about scientific quality and ethical issues. Prof. Carsten Carlberg together with a Luxembourgish team member and co-author published methodological errors that were only corrected in BBA 2011 (culture conditions of primary macrophages yield apoptotic cells). Without informing people involved, they took apoptotic, bad samples to generate results presented in Figure 1B. Doubts were raised about biological triplicates. Ethical issues relative to use of human material were not addressed. One may wonder why University officials did not ask Prof. Carlberg to retract this publication.
Thank you for the infomation, Sepi. Few details of the Luxembourg investigation were made public.
In which manuscript is fig. 1b you are referring to, BBA 1803(11): 1276-1286, 2010?
Also, take a look at Fig. 1A in BBA 1803(11): 1276-1286. Some values are marked as significant (p<0.05 or lower) relative to timepoint zero despite the error bar almost touching the "1" line. SD (standard deviation) only means that there is a 68% chance of the real value being within the range, if I remember correctly. If you double it, it's 95%. I think something is seriously wrong with their statistics.
Excuse me! I intended to address Lulu, not Sepi.
In reply to your question TK, Fig.1B is in manuscript published in BBA 1803 (11) 1276-1286, 2010.