How to do any genetic research you want without getting permission.

Let's say you want to do a market-related study in which you gain entry to one thousand homes representing sets of people defined by the usual variables of income, ethnicity, urban-suburban lifestyle etc. The first thing you do is to ask a few people, real nice like, if you can go through their stuff and take a lot of photographs and notes. Most of them say no, and you perhaps even discover that the one or two who actually agree to this are odd ducks. So you go to Plan B. This involves breaking into the homes when the people are out so you can get your data despite the fact that they don't want to participate. But you get caught and can't do that any more. So, now you are faced with the reality that your research plans are done for.

But wait, there is a way to get similar data without needing any permission from anyone and it is not illegal, and in fact, it could actually be cheaper and easier than your original proposal and, while it may not provide the same exact results, it could even provide BETTER results. Let's call it Plan C.

Plan C involves looking at every iteration of every single Google Street View picture ever taken anywhere in the US at any time. All of them. The vast majority of these photographs will show you nothing, but every here and there, you will get some data. There will be a shot that shows a thing in an open window, or an open door, or an open garage, or being carried into our out of a person's house, being delivered, thrown out on the curb, sold in a garage sale, sitting on the lawn after an explosion or fire, or in use (especially yard and garden implements). This is not the same thing as sampling hundreds of houses once each, looking at all contents. But it is sampling the homes lived in by hundreds of millions of people, and sampling them dozens of times over a few years. That could be some amazing data.

And nobody can stop you from studying this source of information or doing this research. If someone does try to stop you with silly regulations from a University or something, just change into a Journalist. Then you're golden. It would be WRONG to stop a journalist from using this information!

Turns out that this works with genes, but even better. One might want to study the relationship between a putative genetic marker and a possible behavioral thing, like maybe a psychiatric disorder or something, in a genetically bottle necked tribal group somewhere. You can try to get permission to do this, and maybe you'll get it. Maybe you won't get permission but you can certainly steal the genetic data from some place and do the research anyway. But people will get mad at you and you'll not get any more research money.

Or, you can go to Plan C.

Here's how Plan C works with genes. We have a good idea of the distribution of many genetic markers that have to do with geographical patterns over time. (Some people would use the term "race" in that sentence but that's incorrect and unnecessary.) So, something like "East Asian" or "Native South American" or "Central African" or whatever has a list of genetic markers that go with it. Genes are supposed to "independently assort" and act all random and all, but they don't. At the finest level, on chromosomes, genetic markers that are near each other travel together because of "linkage." More importantly, genes move in populations. So, those East Asian genetic markers are not going to get all mixed up with the Central African genetic markers too often. Also, if you have enough samples, some mixing up doesn't matter all that much.

So, if you want to study a disease-related "gene" (allele, a variant of a gene, really), if you think such a thing exists, you can effectively study it in a small population of repressed brown people who, tired of repression and exploitation decided to be totally unfair to you and not give you their blood, by looking at the association of LBP ("little brown people") markers and the alleles of interest. It does not even matter if many of those marker associations are found in totally non LBP people. They are still associated. Genetic lineages are the thing, not human lineages. Humans are merely reasonable approximations of genes, really. Or at least, you can make a case for the associations and get your research funded and published without asking anybody any permissions for anything, just using giant available genetic databases. OK, so, maybe this is not "any genetic research you want." But it is without permission!

This all sounds very nefarious but may not be. Or maybe it is. I'll leave that to the ethicists.

By the way, if you are interesting in a big fight on the Internet about genetic research, ethics, "IRB" permissions, LBP's and science and so on and so forth, I recommend the following, in the order suggested.

First, read this blog post: Is the Havasupai Indian Case a Fairy Tale? by Ricki Lewis (and if you have time, along with it, this related blog post)

DON'T READ THE COMMENTS YET

Then, read this blog post: The Empire Strikes Back by Jonathan Marks

Then, go back to the first blog post (Is the Havasupai Indian Case a Fairy Tale?) and read the comments.

Then, report back here and tell me what you think. Especially about that last comment on the PLOS blog post.

Have a nice evening!

_____________________

Photo Credit: tapasparida via Compfight cc

More like this

An ancestry informative marker set for determining continental origin: validation and extension using human genome diversity panels: Results In this study, genotypes from Human Genome Diversity Panel populations were used to further evaluate a 93 SNP AIM panel, a subset of the 128 AIMS set, for…
Why are brown people so many shades of brown? If you were raised in a South Asian family I'm sure that you've had to deal with the "color" issue somehow. This isn't a cultural blog, so I'm not going to go there, but I do think that the salience of complexion in South Asian culture makes this new…
The post yesterday about the deletion which results in heart disease later in life had some interesting ancestry related material. This makes sense, the genetic maps which I post on now and then ultimately have a medical rationale behind them; eliminate population structure so that you don't have…
A recent study of dog genetics, published in PLoS, seeks to improve the quality of genetic research by better understanding the underlying patterns of genetic variation at the level of specific dog breeds. Sometimes we are interested in the evolutionary relationship between two "species" or…

Just claim to be looking for terrorism. If you look hard enough, you can find it any where. That is the whole point of hyperactive agency detection.

By daedalus2u (not verified) on 21 Aug 2013 #permalink

If I'm understanding this correctly it looks like a member of the Havasupai community came to ASU looking for help with their diabetes epidemic. Dr. Markow agreed to do this screening and obtained funding in part from an org interested in Schizophrenia by planning to also evaluate the samples for sufficient diversity of markers which could later be used for research of Schizophrenia and other genetic disorders. The consent form sounds general enough to have covered both of these topics but perhaps it was not made sufficiently clear to donors that the samples would not be destroyed after the initial research unless they expressly requested that. Additional research into specific disease markers would have required additional consent and participation by the donors and so was not included in the original consent form. Thus the confused and combative exchange in the comments. Whether Dr. Markow intended to do further disease marker research is moot since she would have to have had three things to even consider doing such research (1) a data set with variability sufficient to include groups with both carrier and non-carrier markers for the trait in question and (2) extensive medical histories and medical evaluations of the participants (3) permission from participants. Since Dr. Markow determined that she didn't have (1) there was no reason to move forward with obtaining (2) which she could only have done after obtaining (3).

Later, improved technology allowed another student to revisit the samples and look at variation at a much higher resolution. This sound like it was work to confirm the original assessment of population diversity, repeatably is after all, a core tenant of experimental science. The right to use of the samples for this additional research was contested by the Havasupai, the samples were destroyed, reparations to the community were made, and the student's work was not published.

When people ask her about her intent, it is no doubt extremely frustrating, often data like this is collected and analyzed very broadly and only after this initial investigation can a researcher determine what, if any, further specific investigations can be done. For example, I spent a year working for a USDA germplasm repository and my main job was the collection and extraction of genetic samples from hundreds of stone fruit trees so that we could run basic SSR profiles. The point was to obtain SSR profiles, build a library of data, and to investigate the basic diversity of commercial varieties of prunus and arminaca species with the hope that the data could be used for commercially viable research in the future. There were no consent forms because I was taking leaves from trees, not blood from people, but the methodology seems very similar.

Informed consent is a difficult topic. Most people simply haven't invested enough time in population genetics, statistics, and protocols for preserving research data and materials needed to be truly informed about where and how their contributions may be used. A single page form and two hour interview does seem to be rather spare information for consent on research into a population's genetic diversity as a preliminary investigation to the suitability of the material for further study on genetic markers and disease, but given that the Havasupai approached ASU, and not the other way round, perhaps the simplified forms were a well-intentioned effort to expedite the process towards the results requested by the Havasupai and not a cagey scientist keeping all her options open.

Of course, there is no way for me to be sure of any of this since I haven't seen the consent form, the original research, the research which initiated the investigation, or any of the grant proposals used to obtain funding for work with the samples.

I had another thought! I know, amazing, right?

Anyway my thought was this:
Obviously those that worked with Dr. Markow on the initial study of the Havasupai had read and understood the consent form, but it is not clear to me that any of the later students who were granted access to this material had thoroughly reviewed the consent forms or listened to the interviews.

Shouldn't anyone working with human samples be fully aware of the terms and conditions those samples were obtained under? I'm genuinely curious about best practices here, since all of my background is in plant research, I never gave a second thought to any of these ethics, I never even had to review anything like humane treatment of model organisms since it isn't even an issue for plants.

The interesting part is the scientism of Lewis and Markow. Only the scientific peer-review literature is valid to establish the truth about the affair, not the statements of the witness and the participants of the grants' aproval.
The Hart report, that contains the statements of the participants in the study (and other witness) confirmi, from the beginning, they were researching for schizophrenia.

It seems clear to me that the wet lab blood data was broadly applicable, and I think it is okay to use funding from any interested agency to embark on a study to collect it.
I was ready to leave it at that, but then kept reading.
In order to get consent, you must establish from the donor what the covariate data will be: what disease are you taking it for, regardless if the initial tests are of just of broad applicability?
The covariate data found in the medical histories is the complicating factor, and a study of any disease requires this data.
The wet lab data was collected with the understanding that the covariate data were to be diabetes incidence. Clearly, Dr. Markow had the intention of adding a covariate data set of schizophrenia, if the wet lab data proved useful. It's clear that the wet lab data was not useful, and the covariate data of schizophrenia incidence may not have been there, so she gave up that idea. Now she can truthfully say she never studied schizophrenia.
But the link was already made. Scientifically speaking, Dr Markow had a seemingly okay plan; why seek consent now for the covariate data if it wont work out? As scientists, we see that logically, covariate data can be separate entities; if correlation is not found in parameter A and B, maybe there is between A and C. Just get data set A and B now, and we can search for other correlations later. To me, this seems to be what happened.
Why would this be unethical? Because that's not how others perceived it or understood it. Donors were not educated to this idea, and I don't expect donors to need to understand it. The covariate under the donors agreement was diabetes, and they might not have donated has they realized that Dr Markow had other certain other covariate associations in mind.
Is it okay to return to the Havasupai later and say, "We may have found a way to study schizophrenia from your blood samples, may we please search you medical history for schizophrenia data?" It seems such a benign request and with the best intentions.
We should never place the burden of understanding this subtlety on the donor. Due to socio-economic conditions of free-will donating people, the burden of leading them through the process falls in the researcher to explaining in excruciating plain language.

By Thad Harroun (not verified) on 22 Aug 2013 #permalink