12

Recently my lab published a paper that uses a mouse line from another lab.

Without going into too much detail since not everyone here is a Biologist, this other lab has a mouse with a reporter gene in it for hetero/euchromatin formation (DNA "openness"). This took them a long time (and a lot of work) to create. Basically, if the chromatin structure changes significantly in the mouse, this reporter gene is activated. Tens of thousands of these mice are then randomly mutated, and if a mutation happens to cause a big change in the chromatin openness, their reporter gene flashes up, and the mouse is given an ID and becomes it's own little mouse line to study why/how the DNA openness changed.

My lab was given a few individuals from one of these mutated mice lines - enough to start our own breeding colonies. However, there were some restrictions on what we could do - we can't reverse-engineer the reporter construct, we can't use the reporter construct ourselves for anything, and we obviously can't give the mice to any other labs. We can only examine the phenotype (and find the causative mutation).

Now our work on the phenotype has been published (quite highly in a major journal), and i'm interested to know how Science is supposed to validate such experiments?

I presume my lab still cannot give out the mice we used to generate our data to other labs. I presume that the lab we got our mice from is under no obligation to give out their mice either - and they may not want to since it would contain their reporter.

So I presume this means the results cannot be validated by anyone else without them going through the whole process of creating a new mutant mouse line first (and even then, it doesn't speak to the validity of the data on our mice).

  • 3
    What prevents other labs from getting the mice the same way yours did (i.e. from the lab that initially bred them)? – Cape Code Feb 8 '16 at 14:01
  • They would have to ask, and the final say would be up to the whims of the person who answers the phone. Of course they might say yes - but the risks for that lab would seem to outweigh the pros now that a publication has already been made for these mice. Really i'm curious to know if their exists a precedent for some sort of legal or contractual tool (perhaps designed by the grant-funding body or journal as a condition for money/publication) that the mice must be accessible upon request. – Wetlab Walter Feb 8 '16 at 14:21
  • 1
    Interesting, how would you propose to fund the thing? You can't request a lab to keep breeding mice indefinitely for free just in case some other lab, someday, decides to replicate the experiments. – Cape Code Feb 8 '16 at 16:32
  • Very good point - I have no idea :D Ideally validation - if it was a requirement - would be done as soon as the results were first published, and funded by the same people who funded the publication. But honestly I have no idea what the process for validation is like in science. I don't even know if it's possible to secure funding specifically for repeating an existing/published experiment. My whole question assumes that there is a process for validation. If there is not, it doesn't even make sense to ask how validation works in instance XYZ. – Wetlab Walter Feb 8 '16 at 16:59
  • 2
    When the original lab created the mouse line, did they publish an article on that? If they did, I would expect them to provide enough data for another bio group to reproduce this line. If they didn't, I would expect them to do this sooner or later, as soon as they milk enough profit from their mice. – svavil Feb 8 '16 at 20:07
10

You've published the results in a prestigious journal. In most of those journals, that means you've promised to make those mice available to other researchers. For example, Nature says

An inherent principle of publication is that others should be able to replicate and build upon the authors' published claims. A condition of publication in a Nature journal is that authors are required to make materials, data, code, and associated protocols promptly available to readers without undue qualifications.

(Emphasis in the original)

Science says

After publication, all data and materials necessary to understand, assess, and extend the conclusions of the manuscript must be available to any reader of Science.

Similar verbiage is in most high-quality journals. Typically, you agree to those conditions when you submit, or when your paper is accepted, by agreeing with and/or signing forms.

So you may not have known it, but you have already promised to make those mice available. Do you feel that you can't do that? Then you shouldn't have agreed to do so when you published. If someone requests the mice, and you refuse in spite of your agreement, you may end up being blackballed by the publisher and may never publish with them again. Conceivably -- though I haven't heard of cases reaching this point -- your article could be retracted over your objections, or you could face an academic dishonesty investigation.

(The same applies to the original researchers who made the mice, of course, and in practice most researchers would consider the onus to be on them, not you, to provide the mice. But that's a pragmatic thing. As I say, you've most likely already explicitly guaranteed that you will make the mice available on request, and it's your problem to make good your promise.)

Edit to point to an even more explicit explanation from the NIH, in the Grants and Funding section, Frequently Asked Questions: Sharing of Model Organism and Related Resources:

What is NIH policy regarding the distribution and sharing of mutant strains of model organisms created with NIH funds?

The NIH expects that new, genetically modified model organisms and related resources generated with the aid of NIH funding will be distributed and shared with the scientific community in a timely way, generally at least upon publication of the primary results announcing the development of the genetically modified model organisms. Investigators submitting an NIH application (including competing renewals) are expected to include a concise plan addressing the timely distribution of organisms and resources, unless the proposed research will not generate new model organisms and related resources.

Question 9 is particularly relevant here:

... I don’t want to share my novel animal strains or the reagents used to make them. Can I be forced to do so?

Sharing of research resources is a very important NIH policy and is included as a term and condition of your award.

(My emphasis)

So not only have you agreed to make the mice available when you agreed to the publication terms, if your work was NIH-funded, you also already promised to make them available, even if you don't want to.

That section notes that there are possible reasons for not sharing, and the NIH will consider them, but they must be explicitly stated in the grant (and putting that in your grant will make it less likely to be funded). If it was not explicitly stated in the grant originally, this is not an acceptable excuse.

Other granting agencies have similar policies.

  • 4
    I do not think that such publication agreements obligate researchers to give away their materials. If I build a laser with certain never-before-achieved properties and publish a paper on it, am I committing to giving away my laser to anyone who asks? No, I'm (at most) committing to telling people what I did to build the laser. – Tom Church Feb 8 '16 at 20:19
  • 2
    @TomChurch Agreed, the mice here are clearly physical property that should not be required to be given away. However, the scientific procedure to produce the reporter gene could not be witheld - other labs cannot be forbidden to reverse-engineer it and, in fact, should be provided with the explicit procedures used to replicate it themselves. The IP of how each step in the result was achieved cannot be made secret. I think the meaning of "material" here is entirely relating to documentation and information - things that can be copied and transmitted to third parties. – J... Feb 8 '16 at 20:30
  • 8
    @iayork Required is that "After publication, all data and materials necessary to understand, assess, and extend the conclusions of the manuscript must be available[.]" Are researchers required to provide microscopes? Mouse cages? Tuition for advanced degrees in biology and genetics? Researchers clearly can't be on the hook for all the materials required to reproduce research. They're required to provide what's needed to "understand, assess, and extend" the research. I think extend is the most flexible of those, but clearly researchers needn't become funding agencies in order to publish. – Joshua Taylor Feb 8 '16 at 20:55
  • 3
    Iayork, I want to replicate some of your work, I need you to give me your laptop urgently. – Cape Code Feb 10 '16 at 8:46
  • 3
    Coincidentally, with a half hour of you posting that comment I got a request for a set of reagents my lab published last year. I get such requests fairly often (including for transgenic mice I made). I always do provide the reagents. So do the labs I was associated with before becoming a PI. So have 90% of the labs I've requested reagents (including mice) from. I'm sorry you have such a parochial and bitter view of the scientific world, but your view is the minority, and becoming increasingly so. – iayork Feb 10 '16 at 19:06
0

You answer your own question. If your method is a secret, protected by copyright, or otherwise proprietary, others cannot repeat your experiment, by definition. That is the entire point of protecting the special mice in the first place, to restrict who benefits from them.

And therefore no one can verify your method. They could use a completely different method to prove the same result, but no one who does not own these proprietary mice can directly verity your data.

  • But can't someone force the owners of the mice to give them up for validation purposes? Can't the tax-payers who paid for our/their grant, or the readers of the journal article, or someone? Instinctively i feel you are right, even thought this is the answer I really didn't want to receive :/ – Wetlab Walter Feb 8 '16 at 17:25
  • Well, undoubtedly it would be very bad form for the lab to lend out mice to allow you publish a paper and then hold them back if someone wanted to verify this paper. As some might argue, what makes a scientific proof is the ability of it to be repeated. But, property law is almost certainly on their side, and it is very unlikely anyone could force them to do anything even if they were willing to go to court over the issue. And the lab is far more likely to bulk at lending out the mice to some other scientists they do not know, possibly on the other side of the globe. – Jonathon Feb 8 '16 at 17:33
  • 2
    @JonathonWisnoski I'm one of those who might argue that. If your work can't be replicated, it isn't really science. It's ultimately no different from any other argument from authority. – Monty Harder Feb 8 '16 at 20:33
  • 1
    @JonathonWisnoski Balk at, not bulk at. – JAB Feb 8 '16 at 21:35
  • 1
    @CapeCode The mice are part of the method, and are proprietary. But it's a generic situation, they might as well of just labeled something a trade secret and refused to disclose a key step. – Jonathon Feb 8 '16 at 21:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.