What’s the worst thing that can happen to a scientist apart from having to give up one’s career because of funding problems? Stem cell researcher Haruko Obokata would probably have a good answer for that. She’s been in the media since the beginning of the year: first as female scientist role-model with a spectacular discovery, then as outrageous fraud and national disgrace.
For all of you who’ve missed the headlines (if you haven’t, just skip this paragraph) let me summarise what happened: In January, Obokata and her team published two papers in the prestigious journal Nature, announcing that they had found a very simple method to convert human immune cells into stem cells, mainly just by growing them under acidic conditions. Only a week’s work, the authors claimed, was enough to produce a colony of pluripotent stem cells (or STAP cells), something that with standard iPS methods takes a month and involves several culturing steps. The protocol was so simple that it would have allowed any lab with standard equipment to produce stem cells in no time. After much initial excitement several scientists tried to reproduce the results of the – seemingly foolproof – new method, one of them even in a live blog – they all failed. Meanwhile, a number of discrepancies were found in the papers by attentive readers and published online. A formal investigation by the RIKEN insititute, where the main work was completed, ensued and found Obokata guilty of misconduct on the 1st of April. No april fools’ joke intended. Find the more detailed story here.
Yesterday, the scientific journal Science announced that Obokata had finally agreed to retract both papers in Nature that reported her work. Which basically means resignation and her confession that the papers are indeed flawed. She held up until the very last, but in the end with all her supporters lost and the results of the genetic testing on the table which confirmed that the STAP cells were not produced from the mouse strain she indicated, she had run out of options.
Why did they do it?
Probably everyone who’s heard of the issue has asked the same question: why did she do it? How could she think she would get away with it? A researcher cannot simply throw a paradigm-changing proposition out there and not expect peers in the field to scrutinise it until they’ve succeeded in reproducing the data. Cheating scientist are likely to get caught pretty quickly, in particular if the topic is relevant to the field (which it will be if published by Nature). If STAP would have been real it would have revolutionised the whole stem cell field, and that’s clearly what it was intended to do.
Perhaps Obokata really believed that her method was working – even though she was lacking final proof? Other labs have reported artifact green Oct-4 staining in their cells – autofluorescence – which means these cells seem to express stem cell markers when in reality they don’t. Autofluorescence is a common problem in light microscopy. Every scientist knows and despises it. It is unlikely that Obokata could have been deluded by such a basic interference. And again, the manipulated images indicate a deliberate attempt to portray her findings in a different light. Paraphrasing whole sections of a PhD thesis also does not signify good scientific practice. But why the cheating on such a massive, obvious scale?
I already pointed out the worst thing that can happen to a scientist: to be forced to give up the career because of funding issues. And some researchers go through great lenghts to make sure that doesn’t happen. Finishing a PhD project without a first author publication puts an end to their careers for most scientists. In Obokata’s case she did have 2 first author publications in lower journals in the year she obtained her degree in engineering (2011). However, her thesis project was unrelated to the STAP work which she reportedly came up with at Harvard, where she spent two years in Charles Vacanti’s lab, before finishing her PhD in Japan. The STAP ideas probably won her the position as a guest researcher at RIKEN and, in 2013, the appointment as group leader in Cellular Reprogramming. Having worked on an allegedly revolutionising discovery for 5 years with nothing to show in the end would have been a humiliating tragedy, to say the least. It might certainly have endangered her position. However, the mistake was for the researcher and collaborators to take on such a high risk project in the first place, and the institute to support it without checking whether accurate scientific practice was adhered to.
Perhaps this was how it all started for Obokata. Perhaps it was just a simple, tempting idea during a PhD project that went out of control because safety mechanisms were not in place to prevent her from investing too much time into a concept based on wrong assumptions. We will probably never know.
Does modern science encourage people to cheat?
To focus on unreasonable goals is not a rarity in research, especially not during early career stages. It is part of the training to learn to distinguish between studies that will result in new findings and those which won’t. But usually a PhD adviser or group leader is at hand who can point out misguided approaches or mistakes and these projects will not be carried over into senior positions. RIKEN finds harsh words for Obokata in its formal research paper investigation. Their conclusion is that “she sorely lacks, not only a sense of research ethics, but also integrity and humility as a scientific researcher”. Fact is, as a PhD student you hardly get these values taught these days. Research ethics is not on a typical student’s To Do list and it is certainly not asked for in the respective transferable skills logs of most PhD programmes. Although many of us initially join “to pursue the truth”, reality sinks in quickly. The main request you’ll hear from your PI is not “Have you done your work properly?” but “Where’s the new data?” and “How are you coming along with the paper?”. If you’re lucky enough to have a direct competitor who might scoop you any minute then there’s also a significant amount of time pressure involved to get your research published. You need more controls? Your new transgenic mouse strain is not ready yet? Well, that’s your problem. The temptation to just make up the data you’re missing is high, especially since there’s so much at stake. Today’s science system does not reward the diligent and ethical – it rewards the fast and blatant, just like any other business.
I’m not looking for excuses. There certainly are none for cases like Obokata. They undermine the trust in science and do a lot of damage. But it’s a bit too easy to put her on display as the main scapegoat and blame her behaviour solely on a “lack of morals”. I do not know her. But she might have been as enthusiastic and naive as any PhD student when she started her journey into research. There are more sides to this story and one of them is certainly a flawed environment that has allowed this misconduct to happen. Morally flawed scientists need to be identified and sifted out early on and the importance of ethical behaviour has to be taught. Some European framework programmes, for instance, organise compulsory Bioethics courses for all graduate students they fund. Two sessions I attended were held by Swiss bioethicist Alexandre Mauron and I can highly recommend them. Although scientific misconduct is not the primary concern of most bioethics courses, which typically focus on ethical issues brought about by scientific progress, it does fall under its general scope and should be emphasised more. Of course this will not deter a truly amoral, attention-seeking scientist to cheat, but for the rest of the field it can act as a reminder and perhaps inspiration.
Equally, no-one ever gave a sufficient explanation as to why Obokata’s research was finally published by Nature after being rejected the first time. Or why neither her co-authors nor the peer reviewers found any mistakes. How are we all going to respond to the next big announcement in the cell reprogramming field? Even Shinya Yamanaka, who pioneered iPS cell research and was awarded the nobel prize for it, talked highly of Obokata’s research after it had just been published. He obviously trusted Nature’s peer review system, just like everyone else.