For reasons I don’t entirely understand, this post is trending high on Google searches for ‘chemistry fraud’. I’d actually like to do a follow up on some of these cases, so if you’re interested in seeing that, have other cases you’d like listed, or know any of the participants, please contact me in the comments or at: chemistrystatistics AT gmail DOT com
- verpa , 10/20/2010
After reading about the IUCr scandal with some 70 structures invalidated followed by another one this month for another set, it got me thinking about the chemistry scandals that have come to light in the past few years. It seems as though the number of massive frauds is increasing … or are they just getting becoming more public? A quick review of some of the biggies ( only chemistry, mind you ):
- 1994 – Guido Zandel
Claimed to have discovered an enantioselective preference in some reactions under a static magnetic field. Took until 1996 for the paper to be retracted. Uncovered from third party failing to reproduce results. This is one where the system worked but it still took several years and wasted many people’s time. Total submissions withdrawn: 1. - 2000 – Peter Chen
I feel bad using his name on this one, since he seems to have tried to deal with it properly. Apparently one of his graduate or post-docs falsified data used in their thesis, and later in several publications. Chen realized it when he couldn’t reproduce his own group’s results, and requested a committee investigation. Then things apparently got ugly. Uncovered by his own efforts. Total submissions withdrawn: 3. - 1998-2002 – Hendrik Schön
Author of an insane number of papers and ‘breakthroughs’ at Bell Labs. Took several years before anyone noticed that he was submitting the exact same data repeatedly. Also somehow managed to lose all his lab data and notebooks. Uncovered from a third party noticing duplicate data. Total submissions withdrawn: 21. - 2003-2007 – Pattium Chiranjeevi
Again an author of an insane number of papers. Allegations include recruiting graduate students to scour old journal articles to pirate, changing small features of an article and publishing in a different journal. Uncovered from third party noticing their own results being plagiarized. C&EN news seems to think this a good place to shill for a plagiarism detection system … too soon man, too soon. Total submissions withdrawn: 70+.
If you don’t mind the profanities, The Curious Wavefunction has some hilarious comments in their discussion the C&E news story. - 2007-2009 – Crystal Structure Frauds
Drs. Hua Zong and Tao Liu were found to have submitted scores of fraudulent structures to IUCr. In many cases, these structures were chemically impossible and appear to have been created by manual edits of existing published structures. IUCr has published a very detailed editorial about how the fraud was committed. Now we have the UAB retraction as well, and you have to imagine there’s more in the database to be found. Discovered while running shakedown tests on new software against ‘real’ data. Total submissions withdrawn: 70 + ?? … oh god, I just give up.
With the most recent batch of frauds coming out of China, it’s easy to blame China’s research-or-starve model ( literal publish or perish ) as the cause. But, as Nature continues with another piece, that’s too simple of a view.
We need to submit journal articles to the same kind of plagiarism scrutiny that we force students to undergo. Reviewers can’t be expected to keep a significant fraction of an exploding field of knowledge in their heads. We need to let the computers score new manuscripts/data for duplication across all the journals, then include those scores when the manuscripts are sent to the reviewers. I’m not saying we should pull the human from the loop, but the computer should be in there too. Without this sort of check, the databases are rotting from the inside.
We should also have open commenting on all journal articles once published … not the anemic ghetto of a journal club that JACS is doing to ‘test the waters’. As far as I can tell, no one takes that ‘club’ seriously, as it’s obvious JACS isn’t committed to it. I wouldn’t mind having to defend my papers against the occasional troll if I could see easily that three people other people had failed to get a synthesis working, saving myself a day or two of work.
How often have you heard a fellow chemist say … “I don’t trust anything that’s not in [ JACS, Angewandte, Organometallics ]” or “I won’t even read Tett. Letters because it’s full of ISHTAR: Irreproducible Shit That Aggravates Readers“?
Is the network of trust between chemists breaking down?
Where do you see this heading?
“We should also have open commenting on all journal articles once published … not the anemic ghetto of a journal club that JACS is doing to ‘test the waters’. As far as I can tell, no one takes that ‘club’ seriously, as it’s obvious JACS isn’t committed to it. I wouldn’t mind having to defend my papers against the occasional troll if I could see easily that three people other people had failed to get a synthesis working, saving myself a day or two of work.”
You might check out Chempedia Lab – it’s designed with this idea in mind:
http://lab.chempedia.com/
Believe me, I keep an eye on chempedia lab ( and have posted a time or two). I’m a big fan of the StackExchange community model. But, and maybe I’m too used to stackoverflow, but isn’t the idea with StackExchange more about getting quick answers to concrete questions?
Might be good for a ‘can anyone reproduce this paper’s data?’ type question, but I don’t see anyone using it that way already.
“But … but isn’t the idea with StackExchange more about getting quick answers to concrete questions?”
That’s the intended purpose of Chempedia Lab, although few are using it that way. I’m hoping that Chemists like you can show the way by doing.
I invite you to post a specific “can anyone reproduce this paper’s data?” question. It shouldn’t be hard to come up with one or two examples.
The worst that can happen is nobody responds. Even that seems unlikely because the number of questions with zero answers is pretty small:
http://lab.chempedia.com/unanswered
OTOH – you might actually get an answer you wouldn’t have gotten otherwise, or you might save someone else a day or two of work.
Unfortunately I’m not a practicing chemist anymore, thanks to the downturn, which means my acs subscriptions went away. Hence the focus on free modeling software
I made http://www.chemfeeds.com so people can stay on top of the literature and comment if they wish. Hardly anyone does, but the community isn’t large. Voting is broken at the moment though.
That’s exactly the sort of thing I wish we could get integrated directly with the journals.
Have you ever considered pitching it to the NSF? I doubt without the backing of the NSF the journals would ever really get on board.
Too bad you and Egon aren’t collaborating to get chemfeeds and chemical blogspace’s journal comments working together. A bit of pubsubhubbub might help build ‘virtual’ communities combining your projects.
Also would like to say, that I love the captcha for chemfeed commenting.
Egon and the other guys always approach things from an informatics perspective. I always tend to approach things from a community building perspective.
I haven’t looked at getting NSF funding or participation, partly because http://www.chemicalforums.com got very popular over time and I’m patient. Also, I tend to believe that the chemists who will be using http://www.chemfeeds.com the most are still in high school, so the audience is inherently small. When I approached some ACS people about allowing comments on articles they thought it was too dicey of a subject to manage. I know Nature China allows comments on their articles and it has been well received by everyone.
Do you plan on covering any house science committees this week? Be best if we didn’t double dip.
P.S. Thanks for enjoying my robot captcha though.
No committees for me I think, if I do, it will be old ones. I’m still trying to catch up since I discovered that resource through you. And anyway, 3/4 of my meager readership comes through you anyway.
Yup to the ‘ACS people about allowing comments on articles they thought it was too dicey’, that was the exact fear I heard when I talked with older chemists in grad school about technological developments a time or two. The worrying thing for me was, when I talked to undergrads and fellow grad students / young profs about the same idea, I got basically the same distribution. The techies were all for it, but there was still a large section that was ambivalent … the ‘we’ve never needed it before and our respected peers don’t think it’s a good idea’ / ‘journals are not myspace’.
“Egon and the other guys always approach things from an informatics perspective. I always tend to approach things from a community building perspective.”
@Mitch – did you post a more detailed writeup of what you mean by this? I remember you writing similar before, but not seeing anything in detail to explain what you mean.
The only reason I ask is that I think I know what you mean (creating – or changing – culture is harder than creating software), but it might be helpful to dive deeper into it.
@Rich I simply want to create platforms and websites where chemists like to play. Things like RSS, semantics, wiki, fill in the buzz word, comes a distant second.
For instance, I’ve spent a year cultivating a chemistry community on top of the Reddit platform, a social bookmarking website. It has good activity with over 2100 members.
http://www.reddit.com/r/chemistry/
That would be an example of my focus on community development and not chemical informatics.
Changing the culture has never been my goal, or at least not anymore. I have a more, if you make it – they will come philosophy. I aim for the young generation, the older chemists can go about their business.
@Mitch, if I can jump in for a sec, I’d be very interested to see a full post about your experiences with community building and what led you to the last paragraph in your above post.
Sounds like @Rich might be interested too.
Do you think that students brought up with your tools will be invested enough to hold on to them through the grad student apprenticeship, where a fraction of older profs may say ‘those were nice toys when you were younger, but in my group I control the communication with the outside world’ until they are running their own lab?
Or do I, like the informatics people sometimes, have too much a young turk attitude?
You almost make my point for me about the differences I hinted at. You said, “Do you think that students brought up with your tools will be invested enough to hold on to them through the grad student apprenticeship.”
The question really should be, “are the students that have been brought up in online chemistry communities invested enough in that community to hold onto them through the grad student apprenticeship?” It is the community’s job to retain the younglings whether it is a Q&A community or a social bookmarking community, my job is to make sure the platforms look shiny and the servers are maintained and paid for.
Since you asked me to elaborate on my previous last paragraph and as I really don’t want to make a post on it, I’ll touch on it again. Professors have their own chemical networks that they either developed or were brought into by others. I consider professors as very efficient networkers, meaning that the benefits of entering an online network don’t balance out the loss of time they could have used in cultivating their established network. A lot of chemists (grad students and younger) don’t have access to the professor networks so I try to provide them alternative means to tap a global online chemical network, as those are the individuals that would be most in need of networking. Since I have made chemicalforums.com I have seen the high school students I helped go to college as chemistry majors. I’ve then seen them finish undergrad and either take jobs in industry or enter graduate school. Once in graduate school they seem to like chemistry-blog.com more as they can relate with it and some have even written for the blog. The more senior graduate students seem to love chemfeeds.com because it makes staying on top of literature easier. So I am not worried about engaging professors, these students will be the professors soon enough. Plus, I tend to assume if the online chemistry communities exploded in activity and networking the professors will come — but hell, there really is no need for them on a long enough timeline.
Mitch
ACS Nano editor on the Chinese fraud and other issues you seem to like.
http://pubs.acs.org/doi/abs/10.1021/nn100182y
Mitch
Thanks for the article, I liked the quote:
“While the h-index does have some utility and convenience, the dangers of … unhealthy consequences of their frequent consumption, just like simple sugars, need to be remembered very well … Who knows what may lead to the next breakthrough in science or technology, as has happened many times in the past?”
Echos of Thomas Kuhn.
Plus it was just funny to see so many analogies packed into the final paragraph.
У меня почему то кодировка не верно отображается, одни каракули, как это исправить?
К сожалению, я не владеют русским языком.
[...] A Chemistry Fraud Roundup February 2010 17 comments Possibly related posts: (automatically generated)#1 Page and Top 3 Posts In 2009 [...]