uBiome Is Full Of Shit

uBiome claims that they are exempt from IRB review because their “primary purpose” is to provide a service to private parties for a fee, and that their subsequent analysis of the data collected is an incidental “meta-analysis”:

However, IRB approval is not required for us to provide our primary service of microbiome compositional analysis and interpretation for private parties. Thus, our sample collection is part of a service and our research study is a meta-analysis of de-identified data, which is technically exempt from IRB.

This is disingenuous post hoc rationalization of their failure to obtain up-front IRB review of their human subjects research project before recruiting and enlisting study participants:

H+: You are looking for funding via Indiegogo. Can you tell me a bit about the project you are looking to fund specifically and describe what funders will receive? Why did you choose Indiegogo over other services, i.e. Kickstarter?

Jessica: We are looking to crowdfund our citizen science project at www.indiegogo.com/ubiome. For $79, you can sequence your GI tract; for $235, you can sequence the GI tract three times; for $272 you can sequence all five sites: mouth, ears, nose, GI tract, and genitals. Here’s how it works: You pledge and we send you a sample kit. You swipe the sample brush across the corresponding sample site. The cells are lysed in solution and sent to our laboratory for processing. Once we’ve sequenced your sample, we’ll send a login to our website where you can visualize and understand your data. As a uBiome community member you find out about our latest discoveries first, participate in ongoing citizen science projects, and suggest new questions that we can address together. We chose Indiegogo because they are international, flexible and have great customer service. However, we hope that a lot more citizen science projects will come online, on any platform.

[Emphasis added.] In case it wasn’t obvious enough already, these public statements by Jessica Richman, a uBiome principal, make it absolutely clear that a primary purpose of the uBiome project is scientific research inquiry, and is far from incidental. And also, the claim that study participants are not recruited or enlisted until the kits are used to collect tissue/fluid samples and returned is also a load of horseshit: note all the uses in this public statement of “you” to refer to the same person paying the fee, receiving the kit, sending in the tissue/fluid samples, and thereby contributing thir microbiome sequences to uBiome’s “discoveries”.


  1. says

    What I find utterly fascinating about this whole debacle is the unwillingness to say, “We made a mistake. Let’s stop, take a look at where we went wrong, and figure out how to do it right from here on out.”

    Is that just not done in the world of start-ups?

  2. says

    Is that just not done in the world of start-ups?

    1) It takes a certain amount of hubris to start a business
    2) It often takes a certain willingness to be loose with the truth to start a business (especially if it involves pitching venture capitalists or investors on a business plan)
    3) Startups have boards of directors and if, in #2, the founder omitted to mention that they overlooked something extremely important, the board of directors might make note of their stupidity

  3. Reverend PJ says

    I seem to recall that an IRB makes the determination that a project is exempt not the investigator.

  4. fuckesatonne says

    Chaaaaaaaaaaaaaaaaaaaaarge! Generale PhysiProffe rides to the rescue! With his trusty sidekick Sancho Gregory In Seattle just steps behind! Woooheeeooooo!

    Y’all should really back the fucke off.


    They solicit your goddamn samples and analyze them for you, WITHOUT intending to do any research on them.

    THEN, once they have your sample and a million others, they sit around scratching their butts, trying to think of some cool question they could ask with all these samples they collected. Then they formulate an experiment (which will be funded by all this money you just paid them), get it approved by an IRB, and send you an e-mail asking if pretty please, can they do some research on your sample. And if you say no, they throw away your data and don’t look at it.

    Now, maybe that’s not how YOU would go about doing a human subjects experiment. But you aren’t a small start-up full of energetic young people who want to crowd-fund and haven’t thought through every last niggling fucking detail, they way you have to do in your stooopid grants. No, you’re a patheticke, stodgy, uninspired scientist who has to convince a bunch of other patheticke, stodgy, uninspired scientists to give you a good grant score to fund your patheticke, stodgy, uninspired science. And you just can’t STAND it that these kids didn’t… wait for it… FILE A PIECE OF PAPER.

    Re. Fucking. Lax. No one has gotten hurt here, and now that they’ve promised to file your precious piece of paper, no one will.

  5. bsci says

    Then they formulate an experiment (which will be funded by all this money you just paid them), get it approved by an IRB, and send you an e-mail asking if pretty please, can they do some research on your sample. And if you say no, they throw away your data and don’t look at it.

    This is actually not what they are currently claiming. Here is what is actually being stated: They solicit your samples and analyze them for you. Even though they intend to use these data for research, the actual data collection is considered a private transaction. Once they have received your data, they can then anonymize your samples and do whatever research they want on them without ever obtaining a consent form. They might include a consent form as a formality, but according to the current stance of uBiome, it is completely unnecessary.

  6. infraredeyes says

    OK. Suppose they do a meta analysis on all this data that they just happen to have accumulated while offering a service to the public. Suppose they discover (and I am making this up) a strong correlation between a subspecies of gut flora and, say, resistance to colon cancer. What do they do with this knowledge?

    I would think that the ethical thing to do would be either (a) publish it openly or (b) patent it, then open up the patent for universal use (a thing one can do).

    I haven’t read their fine print, but it sounds like familiar business plan:

    Build huge biomic database

  7. cackleofrad says

    Interesting debate which influenced me to think about the need for IRB approval for some of my work. Word from the IRB at uni (which has pretty big biomed campus): no name associated with sample, no approval needed. My work is fine, but I do not think it will be so with the UBiome folks.

  8. DrugMonkey says

    geez, if you weren’t so mean and all, then they would totally agree that they made a mistake. But with your uncivil tone what could they possibly so but respond badly?

  9. DrLizzyMoore says

    Wait. Did someone just actually say give them a pass bc they are young and haven’t thought through all the details???? Seriously!!!??? Ethics, people. It does not take too much time to be ethical.

  10. DrugMonkey says

    Btw, is there much chance that this unrepresentative sampling of buttflora is going to get us anywhere? What sort of numbers are they going to have to exceed to do better than a good representative sampling design?

  11. says

    CackleofRad: I will admit that I am very pleased to see this as an outcome. I think it is important, even if the work ends up being exempt, that we ask the question. How does my work impact society and how are the people that interact with it as subjects potentially impacted by their participation? That really is one of the first questions we should be asking when we initiate a study. Every time.

    DM: I was academically very curious about the buttflora issue myself. How did they come to the conclusion that with 500 samples they could address depression and arthritis and with 50,000 they could address “neonatal progression” (whatever that means) and cervical cancer. I’d love to know the details of that sample size calculation and to know exactly what questions they think they can address.

    And then there are the statistical considerations around multiple posthoc comparisons…

  12. fuckesatonne says

    “Did someone just actually say give them a pass bc they are young and haven’t thought through all the details????”

    No, no one suggested to “give them a pass”. I suggested that you not make such a big fucking deal (OMFG! Report them to the FDA! And to the dogcatcher too!) about what essentially amounts to a failure to file a stupid fucking piece of paper. Now that they’ve promised to do so, you can go to sleep tonight assured that uBiome will not perpetrate another Tuskegee on the universe.

    All the hoopla here really makes me wonder about academia. Do academics really have nothing better to do than sit around snarking at each other for failure to cross all the Ts and dot all the Is? Damn. I thought scientists were at the cutting edge, head in the clouds, preoccupied with gaining new knowledge at the expense of almost everything else. But it turns out y’all are just a bunch of bureaucrats in disguise.

    Vogons. Bah.

  13. qaz says

    I love the naivete implied by thinking that the issue is that they “just need to file a piece of paper.” The point of the IRB is a process to ensure that a lot of subtle issues about appropriate use of human subjects have been addressed. It’s not the paper that matters. It’s going through the process. These processes are in place because its really easy to screw this up. (And because they used to screw this up all the time. Even when trying to do it right. Read Rebecca Skloot’s book on Henrietta Lacks. Many of those people were trying to do the right thing. Yes, it’s easy to say Tuskeegee was bad. It’s a lot harder to realize that anonymity matters. And it’s a lot harder than one thinks to really anonymize data.) Its really easy to say “we thought hard about it. It’ll be fine,” but its a lot harder to actually do it right. The reason we have inspectors is because there is a conflict of interest in inspecting oneself. And it has been well established that a company is much more likely to do a good inspection if they think they’re going to get checked by an outside inspection agency. And that’s not even worrying about conflict of interest. (No company is going to say “we did it wrong. close the whole thing down.”) Filing the paper is a means to an end. Getting them to do it right is what’s important.

  14. fuckesatonne says

    There is no naivete. I am 100% in favor of uBiome getting IRB approval. I am 100% in favor of them having done this before they solicited samples, but the fact is that they didn’t do that. What I am not in favor of is crucifying them for their eagerness to get some actual work done and consequent sloppiness. Now they’ve promised to get IRB approval, so there is no need to (literally) make a federal case out of it.


  1. […] Several recent blog posts and a session at Scio13 (discussed here) have addressed ethical issues in citizen science. Ethics in research is taken extremely seriously in academia: every single research project that involves human subjects gets reviewed by an independent committee (an Institutional Review Board, IRB) before it begins. […]

Leave a Reply

Your email address will not be published. Required fields are marked *