The Portuguese cancer researcher Sonia Melo was found guilty of “negligence in handling and presenting data” in her publications by the European life science society EMBO; she and her doctoral advisor Manel Esteller also had to retract a paper (Melo et al, Nature Genetics 41, 365–370, 2009). Despite further image irregularities in his other publications (where Melo was not a co-author), Esteller was tasked by his Spanish research institution IDIBELL to investigate himself.  As Principal Investigator (PI) he cannot be responsible to vouchsafe data integrity in his own papers, decreed IDIBELL leadership (of which he is actually part of).

The French plant scientist Olivier Voinnet still enjoys full institutional protection of his Swiss university ETH Zürich, despite his self-admitted misconduct. However, before he acknowledged to have been excessively manipulating data in his own papers, his subordinate researcher Patrice Dunoyer accepted sole responsibility. When the first retraction (of currently seven) hit, the retraction notice  of Dunoyer et al., Plant Cell 2004 read: “We wish to state that the first author, Patrice Dunoyer, was solely involved in generating the erroneous figure panels”. Dunoyer’s reward for such loyalty: despite a confidential CNRS investigation against him, he was also allowed to remain in his permanent position as group leader.

Most recently, Cell Press was faced with a dilemma. It was contacted by a junior author, Yao-Yun Liang, who admitted to have “manipulated the experiments to achieve predetermined results” in the papers in Cell and Molecular Cell. It is safe to assume this whistleblower also provided solid evidence, since Cell Press issued two Expressions of Concern (here and here). The last author, Xin-Hua Feng from Baylor College of Medicine in USA, denied everything “citing concerns about Liang’s motives and credibility”. Yet, this being Cell of Elsevier, the publisher simply tasked Feng with investigating himself. He was invited to reproduce the flagged experiments elsewhere, presumably followed by some “Voinnetting”, namely to use those to correct his manipulated paper. Unfortunately, I did not succeed reaching out to Liang, also his past collaborators did not know his current whereabouts. Feng, unsurprisingly, did not reply at all.

One can continue listing ad nauseam examples of retractions, corrections and expressions of concern where a junior author was assigned the exclusive blame, while the PI was presented as a hapless victim. It seems labs all over the world are truly infested by ruthless scheming PhD students and postdocs, whose only goal in life is to bite the hand that feeds them. The innocent PI is guilty of nothing more than keeping such snakes at his honest bosom, this is at least how universities and journals like to publicly present the instances of research misconduct. No-one wants reputational damage or loss of funding to hit their faculty, or to lose important contributors of exciting research papers. However, the reality is often somewhat more complicated.

The personal account below is likely to be representative of many cases of research misconduct, but also of lab safety abuses.

Here I present a guest post by “Ernesto”, a young scientist who asked me to keep his identity confidential. Below is his personal account of how he came to commit research misconduct and breach radiation safety regulations, under explicit approval of his supervisor. Certain specific details which would allow direct identification of the persons involved are omitted, though all known to me. The words are “Ernesto”’s own, but I did some editing. Your reader comments will be most welcome to both of us, also any advice you wish to offer “Ernesto”. Especially if you experienced or witnessed something similar.


By “Ernesto”

I successfully defended my thesis in 2015 and I currently hold a PhD in pharmacology and physiology from a Southern European university. In one of my publications, in the journal “Toxicology and Applied Pharmacology”(where I am a first author), there are three panels containing images which were manipulated and do not exist materially in film as such. I am ashamed of having sold my principles for it, even more when it was not really necessary. I performed these manipulations on the original scans of my own films, with the express knowledge of my advisor.

Why did we do it? My PhD scholarship was renewed annually and the fourth year was not awarded to all. The situation got even worse as the general economic situation in the country worsened and all institutions found themselves strapped for cash. Also, my thesis advisor, despite being a PI with his own project, was not employed as permanent staff. He had a temporary contract in 2013, which was due for evaluation that year, as was my scholarship for the fourth and final year. I needed a publication as first author for my evaluation and he needed another publication for his.

Disillusionment and dishonesty

This is how I came to his lab: I got my national equivalent of a Bachelor’s degree in Biology in 2007. I had some experience in lab work due to some volunteering and some free choice credits in genetics and, although I was not specially looking for a professional career in research, I was offered a contract in a plant physiology lab. I had a good relationship with my PI (whom I consider my scientific father) and at that time I was tentatively considering doing my PhD in that lab. Sadly, I chose to leave the group, since the chairs of plant physiology (husband and wife) had an adversarial relationship with the people of the group I was in (briefly put: the chairs had ruled the plant physiology department with an iron fist for about two decades until the people in my group got their own project and financing and started working independently). Since that made the environment pretty uncomfortable to work in, due to daily small actions of “sabotage” engineered to hinder the work of my group, I was looking for a PhD position elsewhere.

As fate seems to have a keen sense of humour, I got a state PhD scholarship in 2009 in a group led by one of the godfathers in the field of Arabidopsis in the country. The day to day conditions in the lab were far worse than those in my previous position. I had to juggle parts of three different projects at the same time, my boss had some control issues (he had a thing with keeping people in the lab until he left at 20:00 and usually panicked when we were working somewhere else in the building). The air in the lab was thick with the smell of burned-out staff (overall, post-docs, but the average pre-doc took about six or seven years to get their thesis done). One had been working on her PhD thesis for several years after her scholarship ended, living off her unemployment benefits.

I was assigned a collection of some 12+ different Arabidopsis mutants to keep track of, so when I discovered by sequencing that one of them was not what it was supposed to be, I reached my breaking point. I bailed out and renounced the scholarship three months in. I was left with a feeling of personal failure and inadequacy, but in the long term it saved my sanity and health. Although the lab usually published on the level of PNAS and Plant Cell, I don’t think it justified those working conditions.

In 2011, I got another scholarship, from a private foundation, to work on a thesis on nephrotoxicity at a national key reference centre for kidney studies. My advisor was the very PI with whose explicit knowledge and approval I manipulated the image data which we then published. My co-advisor was the current chief of the medical service at the hospital which our research institute belonged to, and although he didn’t supervise me on a day by day basis, he was far more supportive.

The images I manipulated were of EMSA (Electrophoretic Migration Shift Assay). This method is used to determine the binding of proteins, usually transcription factors, to DNA under experimental conditions. A DNA fragment with a consensus target sequence of a determined promoter is labeled with radioactive isotopes, in my experience with Phosphorus 32. This probe is mixed with the protein samples of your experiment and loaded onto low concentration polyacrylamide gel. The gel radioactivity signal is detected by exposure to an autoradiography plate. Higher binding of the transcription factors is indicated by higher saturation in their lanes, as more protein retains more probe. Unbound radioactive probe ends at the bottom of the gel. Importantly, besides the control lane, a competed lane must always be loaded: this is the unlabeled DNA probe incubated with a random protein sample beforehand (to saturate the retention of the proteins) and then an equal concentration of the labeled probe is incubated with it. This proves that the proteins themselves are not labeled and that the labeled probe binds to the protein specifically.


The image shows contrast-enhanced original figures in “Ernesto”‘s publication (cropped to prevent direct identification). I inspected those for image irregularities, and what I pointed out with red arrows, was confirmed to me by “Ernesto” as traces of his digital manipulations. -LS

The three manipulated experiments of my paper belong to about a dozen EMSAs I performed during my first year of thesis and there is another twist: I performed these radioactive assays without supervision and even without receiving a radiation safety course.

Safety Issues

The national Nuclear Security Council (NSC) demands that isotope facilities in research centres, public or private, have at least one authorised operator, and in hospitals there is a Resident Radiophysicist responsible for the security of irradiating sources, encapsulated or non-encapsulated, which covers nuclear medicine isotopes as well as x-ray, positron emission and gamma emission equipment. Research staff do not need to have the operator qualification but the NSC requires that the centres organise courses and issue authorisations to the in-house staff or that any in-house staff that works in the facility is supervised by somebody with the proper authorisation.

During my first year of thesis work, the facility staff didn’t organise the course. I had to perform the EMSAs with supervision by one or another of my labmates, but I did at least two or three of those without supervision because “it was critical”. You can imagine for whom. There were no accidents, spills or anything (I pride myself in my “hand” skills), but that doesn’t excuse anything.

The isotope facilities at my institute were originally placed at the 2nd floor of the building. It consisted of a clearly identified area with a warning on the access doors as to the unencapsulated radiation sources. We had beta isotopes, such as we used in our lab, but also alpha and the much more dangerous gamma isotopes, as well as x-ray emitting devices. The problem with the facility was that, although it was to NSC standards in every physical requirement, it was hardly an access-controlled area: the institute belongs to a hospital with public access, and the first and third floors of the building were consulting rooms and other medical attention areas. Every once in a while you could find lost patients wandering where they should not be. Besides, the operating technician for the facility was placed in a room far away from the entrance. If anybody with the knowledge and the intention had wanted to steal an isotope vial…

The situation got better two years ago, when they moved the facility to a set of new underground  labs, which actually used to be an underground parking garage, with only one exit to the surface, a single stairwell. I sincerely doubt there’s been an official work safety inspection of those conditions.

The end of the relationship

I finished my work at that lab burned out and disappointed. I know it’s far too usual in research to have disagreements with the thesis supervisor, but as much as I kept a sense of professionalism, I felt systematically undervalued by mine. It got to the point where I kept my source reading to a minimum because he disregarded my ideas out of hand. Nevertheless, I kept working and putting my effort into the job until my scholarship run out in December of 2014, and continued working without pay the following month of January 2015. At that point I confronted him about writing my thesis because he had no financing for me, despite the promises he had made for two months. Since I had been trying to work out an experiment which did not work at all for some three months, I was also pretty burned out and the situation escalated into a hot-tempered discussion, in which I reminded him of that instance of misconduct on both our parts.

After this, I was able to defend my thesis, but we still are in conflict about my authorship on a paper which contains results from the second part of my thesis. As to the EMSA plates per se, I keep them. I was stupid enough to swallow my principles and did something wrong, but not so stupid as to leave it as blackmail material which my boss could use against me.

The most important thing for me is that I refuse to repeat my past errors, despite how hopeless that may make me to become an established researcher. To date, I am ashamed of my ethical breach, and I wish I had had the moral strength to have refused to participate in it. I hope, if anything, that I am not alone in this mess and that more people will tell you stories in this line with the same sincere intent to become better, as persons and scientists, than the men and women who had exploited them.

14 thoughts on “A personal tale of scientific misconduct

  1. Power corrupts, absolute power corrupts absolutely.
    A very difficult one for a young person and congratulations ‘Ernesto’ on extricating yourself from this and recognising that your PI’s path is not one to be followed. Of course you can now safely flag the papers in Pubpeer, so the community knows where the manipulated data are.


  2. “Ernesto” simply represents one case of a culture of fear, cronyism and domination in (plant) science that has become even more acute in the face of financial restrictions in recent years. Those with power and position will try to hand on to it, tooth and claw. Will “Ernesto” stand up to his PI and correct the image manipulations in Toxicology and Applied Pharmacology (perhaps here the TAP editors could act upon this public knowledge), or will he join the rank and file of so many plant scientists who are actively ignoring errors and manipulation in the literature, at the detriment of the integrity of plant science?


  3. In questioning common assumptions, the study authors determined that scientific misconduct typically, but not always, exacts a personal toll in derailing careers. On the public side, the cost to federal funding sources for retracted research was much lower than they expected.


  4. Leo, you need to be a LOT more careful with your anonymization here!

    Taking just the information you present (journal name, approx’ date range, southern European country, and field of study) and plugging into a targeted PubMed Search, returns 4 papers! From there it’s easy to pull the PDFs and find only one with EMSA data. Knowing the “anonymous” person here is the first author of the paper, I was able to discover their name in about 5 minutes.

    While the story is both intriguing and worrisome, I fear your attempts to provide anonymity fall short.

    Liked by 1 person

    1. Ditto – there are now image search services that can look for portions of an image. People believe that just leaving off the name anonymizes a story. But with 3-5 pieces of data (and the Journal name is an excellent selector) you can narrow down pretty much anyone in the world. Maybe you want to check out the (unfortunately now discontinued) blog 33bits (


  5. What does anyone expect? You make people desperate, they will do desperate things.

    Grinding down grad students and postdocs doesn’t produce “the best” science, it simply heaps undo reward on those who do the grinding down.


  6. Much has been assumed about the private and public damage of scientific misconduct. Yet few have tried to measure the costs to perpetrators and to society.


  7. A Portuguese blogger (former researcher from Institute of Systems and Robotics – University of Coimbra ISR-UC ) is reporting and claiming at that a current tenure professor and principal investigator of ISR-UC has plagiarized several research works and project proposals.

    He started saying that the project proposal of the research grant PTDC/EIA-CCO/108791/2008 from the Portuguese Agency for Science funding was a copy past of several previous publications.

    He provided some preliminary screen-shoots of the approved proposal/grants of a 2010-2012 project:

    -Blog post 01:
    -HD image 01:
    (copy from pag.4, section 1 – Introduction, “Facial Dynamics for Identity Recognition, PhD Thesis Proposal by Toby Collins, 2006, Supervisor: Prof. Robert B. Fisher, University of Edinburgh” available at

    -Blog post 02:
    -HD image 02: (copy past from
    Step by step explanation
    (copy from “Simultaneous Facial Action Tracking and Expression Recognition in the Presence of Head Motion”
    (copy from “Investigating the Dynamics of Facial Expression”
    (copy from “Facial Dynamics for Identity Recognition, PhD Thesis Proposal by Toby Collins”
    (copy from “facial dynamics in biometric identification”

    -Blog post 03:
    -HD image 03:
    (copy from page 37 on, chapter 3 “Facial Dynamics for Identity Recognition, PhD Thesis Proposal by Toby Collins”

    If we go the project website we can see that the text of the homepage website is also a copy and past from the introduction of “Facial Dynamics for Identity Recognition, PhD Thesis Proposal by Toby Collins”

    This former researcher of ISR-UC is also claiming that the same PI has 3 another approved project proposals that were plagiarized.

    and started providing details about the documents he is preparing to send to the scientific committee of Electronic Engineering department



  8. While I agree with Paul that it takes only a few minutes to figure out the paper in question and author name, all the details that allow to identify these so easily seem to be provided directly by “Ernesto”. So yeah, someone did fell short in the attempt to anonymize, but the person is “Ernesto” himself and not Leonid. However, details are so obviously presented that I wonder whether getting exposed wasn’t the first aim after all.

    Liked by 1 person

  9. I heard several stories about supervisors forcing their students to manipulate data. Something which little known outside of their groups and these professors look as prominent scientists with funding and even awards. Some students give up to this pressure, some quit, some find compromises like “Ernesto” an dregret it for the rest of their lifes.
    Rare confrontation was demonstrated in Suchitra Holgersson case. The case of Holgersson also shows that even with strong support from some professors (which is also rare in own turn), the chances to win confrontation are miserable.


  10. If the students/postdocs try to confront them, they call them crazy…..there are students totally desperate….they say also they never ask anything to their mentor, they wait he/she tells something….also the mentors never know anything….

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s