Child abuse rips genes

Again on the pet peeve of flu shots for healthy people and antibacterial soaps - a recent report shows that 98% of influenza A strains in the US, and 100% in Canada, are resistant to the flu drug Tamiflu. Hardly surprising given that we've treated about 50 million people with it, and chances are that most of them were improperly medicated.

* * *

So once again, my boss is making waves in the news, this time with actual science, which is a bit of relief. A study mostly carried out by one of our post-doctoral fellows has been getting a great amount of media attention, as it's a very interesting finding. He shows that suicide victims with a history of abuse carry a higher amount of an alteration to DNA structure (but not sequence) that controls the expression of a gene involved in the stress response (glucocorticoid receptor Nr3c1 promoter methylation in hippocampus, for those keeping track). This alteration may "silence" the gene and have rendered these people more subject to anxiety and other negative effects of stress.

My favorite thing about this study has been the press, particularly the BBC's headline of "Child abuse 'impacts stress gene'." One thing I can't figure out about the BBC is why they put those little single quotation marks around a couple words in about 50% of their headlines: "Staff 'too timid' on child abuse," "'Ethical' stem cell hope," or "Veterans at 'higher suicide risk.'"

What do these little quotes mean? They certainly aren't a direct quotes, as my boss would beat you about the head and shoulders for saying the glucocorticoid receptor is a stress gene. They seem to serve no purpose other than to confuse people who want to use real quotation marks on their headlines for citation purposes.

In any case, people tend to take findings like this WAY too far to support whatever point of view that they have. Take the CBC's angle on the story, for example. We've got a journalist claiming that "in a way... the men were programmed to be more vulnerable to overwhelming feelings of despair," which didn't come from any of the researchers on this study (trust me, I know them all personally). Then we have a pharmacologist saying we should aim to "identify these people and then probably offer them some sort of intervention," which is interesting because 1) we have no indication whatsoever that the methylation changes actually cause the suicides, and 2) it's a real bastard trying to get those samples out of living human brains.

From the blogosphere, I think my favorite quote so far has been something along the lines of how obvious it is that abuse is bad and should we really spend millions on research to prove it? I can hardly claim impartiality here, seeing as how my salary gets paid directly from those grants, but the answer is "yes we should, because it's not obvious at all."

OK, so it's obvious that beating your kids is bad (unless you're Russell Peters), but what isn't obvious is why, on a biological level, child abuse is harmful for development. Is it obvious that DNA methylation patterns along the Nr3c1 promoter would be different in the hippocampus of abused suicide victims? Is it obvious that beating your children is like picking up their DNA and twisting it around? Is it obvious why legislators feel like there has to be some kind of biological correlate in order to make a disease somehow legitimate?

We're on your side, people. Don't accuse us of wasting government funds by proving the "obvious." If it's so obvious, then I'm in the comments section waiting patiently for a scientific explanation of the precise gene-environment interactions leading up to suicide.

Genome or Playstation? Hmmm...

Well, we're still here, so the LHC hasn't destroyed the universe... yet. Good news.

Recently there was some not so good news for people who deposited their genetic information (not that way!) into public databases - there is some concern that this information can be tracked back to individuals. This doesn't seem like such a big deal, and it won't be until one day we apply for life insurance and they tell us, "Well, we found your genome online and it looks like you have a 10-fold elevation in your risk for contracting the hoobajabbies within the next 11.6 years, so we're going to need to increase your premiums." Seems a little science-fiction and Big Brother, until you find out that 23andMe (who I've talked about before) can now "sequence your genome" for the same price as a Playstation 3.

Now, I say sequence in quotation marks because they don't really sequence the whole thing, which so far only Craig Venter can really do, at the cost of tens of millions of US dollars. Before the currency started plummeting. Only two people have a data file containing their whole genomes at this point, and they are Dr. Venter and James Watson, and even what they have is a "best guess" from a computer-guided reassembly of their whole genomes from sequenced fragments.

The As, Cs, Ts, and Gs that make up your DNA are known as nucleotides, and there known places where there are differences in the population. These are known as single nucleotide polymorphisms (SNPs), and we have discovered thousands of them. Some of them are even associated with certain (real) diseases, although these studies represent correlations rather than causes. Still, insurance companies calculate premiums based on actuarial tables - statistical correlations - and I don't want them exploiting a $400 sequencing service to make a couple bucks by nailing people who happen to be born with a couple bad nucleotides.

I am seriously impressed that our legislators have already been thinking about this, but clicking on the government link for information about genetic discrimination legislation gets 404'd, which leads me to be believe that it's not exactly a hot issue these days. Like puffin poop.

Ethical Facelifts

Bit of a change in style. Now accepting applications for cool logos.

It's been a long time since I saw so many articles about scientific ethics in the news in so short a period. Well, since Vioxx, anyways. Not long ago, I alluded to the Swiss as having gone "bugf#$k nuts," by which I meant totally crazy. Since 2004, they have enforced a Gene Technology Law, which stipulates that in Swiss experiments, the dignity of [living things] needs to be considered in research. They are in the press these days because they have upheld the decision as it applies to plants, which of course made everyone go "wuh?"

The first effect of this law is to make science even less feasible economically - a Swiss study involving maternal separation in monkeys is going to court now to find out whether it can proceed or not. Making your scientists go to court is only diverting more funds from the state to fight these legal battles, making science that much more expensive to carry out. This is not a good thing for a country's biotech industry, or their "intellectual economy".

Another thing, as pointed out in this Nature editorial, is that dignity is a slippery concept. They cite an American governmental inquiry on dignity in bioethics, which the boys at Nature seem to think is self-contradictory. And they're probably right, but I don't have the time nor the inclination to go wallowing through a 577-page .pdf file written by various and sundry higly respected thinkers to find out that dignity is tough to talk about meaningfully, because it's tough to define, but nonetheless vitally important.

Don't get me wrong - I'm not getting down on dignity. I'm just saying if you ask a dozen people what it is, you'll get 12 different answers, and that this translates to an important point about how advances in the sciences are going to have to force progress in humanities. Development in science is outstripping the development of laws and ethics to control them.

As a direct result of this, you get what we have in California right now, which I also alluded to in an earlier post. California recently issued cease-and-desist letters to various companies that engage in direct-to-consumer genetic testing. Several companies now will sequence your genome (for a small fee) and compare it to known bits of code that predispose you to heart disease, Alzheimer's, whatever. California's decision to require a physician's consent for the tests seems to be an oddly reactionary biotech policy from a state that is known for creating $100 million funds for stem cell research under the nose of the Bush administration.

The legality of this question, assuming a class-action suit against the State by the companies involved, will hinge upon the legislature in this article. Essentially, the decision should be predicated on whether these tests are diagnostic or not. As the lawyers for the defense know, they are not. Diagnostic tools are predictive - a positive result indicates a medical condition and informs the appropriate treatment. What these companies do is associational - they can look at your bits of DNA code and then say "ooh, people with that bit have a 6-fold increased risk of getting the hoobajabbies after their 65th birthday! You'd better start taking antioxidants!" Which is why I agreed with the CEO of one of these companies in my last post when he implied that their industry was entertainment more than anything else.

I think that ethics would be better served by allowing the tests, and providing an option for subjects to make their anonymous genomes publicly available for research purposes. And instead of requiring a doctor's approval, send a Ph.D. to explain what the test results really mean - three studies found that people with this polymorphism have an increased risk of the hoobajabbies, but this more recent study found no association, and the effect size in the meta-analysis is weak; plus there's this editorial in Archives that just came out saying the stats in all these studies are bullcrap anyways. So it's inconclusive.