Mark Henderson - Genetic determinism

07 September 2013

Interview with

Mark Henderson, Wellcome Trust

Kat -  Now it's time to hear from the winner of the Genetics Society's first ever JBS Haldane lecture - Mark Henderson. Mark is head of communications at the Wellcome Trust, and author of the books 50 Genetics Ideas You Really Need to Know and the Geek Manifesto. He's been recognised for excellence in communicating to the public about genetics, and has just delivered his prize lecture at the British Science Association festival in Newcastle. I met up with him before his talk and asked him to reveal some of the things he planned to say.

Mark H. -   What I'm really going to be trying to frame the lecture around is, what I think is often the biggest problem in the way that genomics gets communicated, which often leads to some misframings I think of some of the social challenges that we face as a result of development particularly in medical genomics.  And the problem, the misframing, it really boils down to this, which is that genomics is usually - in the popular imagination at least - understood as a deterministic science whereas in actual fact, it's much more often a probabilistic one.

Kat -   So, you hear these stories, "It's a gene for this", "It's a gene for fat", "It's a gene for cancer", this and that, where the picture you fill needs to be communicated in a much more subtle way.

Mark H. -   That's exactly right.  What has happened I think it's that largely because some of us that have lowest hanging fruit of genomics were actually very deterministic genes, things like the mutation that causes Huntington's disease.  If you get that mutation, you get that disease.  The recessive mutations that cause cystic fibrosis, if you inherit two of them, you will get cystic fibrosis.  There are a number of well-known genetic conditions that work in exactly that way, and they're the well-known genetic conditions out there.  As a result, there has been this widespread framing of genomics is something that's very deterministic.  You have this gene, you have this mutation in this gene, you will get this disease, inevitably.

Now, that's true in a very small subset of diseases and it's true actually in some of the cases in which genomics is starting to make the swiftest inroads into medicine.  It's true for example of very rare developmental disorders where a mutation causes often severe mental retardation, or something like that, where genomics is having a really big impact already.  It's sometimes true in cancer, although in slightly different way where we're talking about acquired mutations in somatic cells most normally and even there, it's not just one mutation that causes the cancer but usually, a cascade of several mutations that are necessary.  So, there are some cases in which that determinism is correct.

But more broadly, it's not actually the case.  Most of the common conditions that we know about - diabetes, heart disease etc. - are influenced by lots and lots of genes which tend to have very small effects.  That's even more the case when you start to look at what you might describe as more social traits - things such as intelligence or height or obesity, or anything like that.  And so, I think if we're to start to think properly about how and why genomics is going to end up having an effect on some questions for society, we need to start from that point of understanding that for the most part, it is a probabilistic science and not a deterministic one.

Kat -   What changes would you like to see in the way that genetics and genomics is communicated?  What do you think could really make the most impact in bringing about these changes?

Mark H. -   I think we have to get away from this language of 'gene for this' or 'gene for that' as you eluded to earlier, you'll see all the time in newspaper headlines, the fat gene or the binge drinking gene, or things like that, and frankly, there is no fat gene.  There is no binge drinking gene.  Some of these qualities, these traits may have an element of genetic influence, but there'll be many genes that are involved and of course, many environmental factors as well.  I think what tends to happen as a result of this sort of 'gene for' narrative is you get two almost conflicting responses, both of which are wrong.

One is this idea that genomics and medical application of genomics is going to be terrifically simply and that actually, it's going to be dead easy to predict what people are going to die of in 50 years' time by measuring their genome.  So, you're going to be able to tell how intelligent somebody is by looking at their DNA.  That's frankly just not the case.  There will be slow progress, slow incremental progress in applying insights of very specific targeted bits of the genome to medicine as we're seeing already in cancer and rare disease, some infectious diseases.

It's not for the worried well who are going to be able to predict exactly whether they're going to get cancer 20 years in the future.  Frankly, whether that comes around is, I think. quite a moot point.  The other thing actually is that genomics causes all sorts of very complicated ethical problems, and it does cause a number of ethical problems.  But some of them I think overblown because of this idea that everything is a gene for this and gene for that.

Kat -   It feels like it really needs to go hand in hand with better understanding of risk and what does risk actually mean for you as an individual.

Mark H. -   That's absolutely right, and something I think that also flows out of this is that very many people are concerned about, for example, what insurance companies will do with genetic data or what employers might want to do with genetic data.  Are you going to have to take a genetic test before you apply for certain jobs?  That's where I think some of this probabilistic, deterministic stuff really matters.

I think that measuring somebody's DNA is going to be an absolutely appalling way to test their aptitude for any jobs.  It's going to reveal virtually nothing.  But if you want to measure how intelligent somebody is or how conscientious they are or how good their IT skills, or something like that, the idea you're going to be able to do that by measuring somebody's genome is frankly rather ludicrous.  And if we do need to think about laws against genetic discrimination, it's not because genetic discrimination will work.  It's because it won't work and we probably need to save people from themselves.

Kat - That was Mark Henderson from the Wellcome Trust, winner of this year's JBS Haldane prize lecture from the Genetics Society. And I'd like to remind you that the Genetics Society's Autumn meeting, looking at how the information in genes is interpreted to create biological shapes - from limbs to leaves -  will take place at the Royal Society on 7th and 8th November. If you want to go, just register now at

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.