Aug 15 2014

Bad Reporting About Epigenetics

Brad Crouch should be fired. At the very least he should never write a science news article again (well, maybe after remedial education and appropriate penance). At first I thought perhaps he was a general or fluff journalist taken off the dog show beat and asked to cover a science news item, but his byline for The Advertiser (an Australian news outlet) says he is a “medical reporter.” That’s frightening.

I read a lot of bad science news reporting, but rarely does a reporter so thoroughly misrepresent the actual science news – unless there is an obvious ideological agenda, but as far as I can tell this is just pure incompetence.

He is reporting on a review article on epigenetics recently published in Science. The two articles have very little in common, and it’s difficult to see how Crouch arrived at his story other than just making shit up. He begins:

LANDMARK Adelaide research showing that sperm and eggs appear to carry genetic memories of events well before conception, may force a rethink of the evolutionary theories of Charles Darwin, scientists say.

First, the paper is not research, let alone “landmark” research. It is a review article. It’s not even a systematic review or a meta-analysis, which a reporter might be forgiven for calling a “study” – it’s just a discussion of the topic of epigenetics.

What is especially interesting is that the words “evolutionary theory,” “Darwin,” and “Lamarck,” appear nowhere in the review, despite the fact these are the focus of Crouch’s reporting. Further, the word “epigenetics” appears nowhere in Crouch’s reporting, despite the fact that this is the focus of the review article.

Here is what the review actually says:

Transgenerational epigenetic effects interact with conditions at conception to program the developmental trajectory of the embryo and fetus, ultimately affecting the lifetime health of the child. These insights compel us to revise generally held notions to accommodate the prospect that biological parenting commences well before birth, even prior to conception.

Essentially, epigenetics are tweaks to the expression of genes in response to environmental factors. If a mother is, for example, living in times where there is good access to food, epigenetic factors will adapt her children to the current abundance. If she is living in lean times, they will we more adapted to food scarcity.

One biochemical mechanism of epigenetic factors that has been discovered is methylization of base pairs. This does not affect the sequence of genes, but it can affect their expression.

While epigenetics is stil fairly new, and the details are being worked out, so far the experiments show that it is a short term (1 or a few generations) tweak for adaptation to immediate conditions. It does not have any real impact on Darwinian evolution. It certainly does not indicate that the inheritance of acquired characteristics (often referred to as Lamarckism, even though this notion did not originate with Lamarck, was not unique to Lamarck, and was actually a minor aspect of Lamarck’s evolutionary thinking) will now replace Darwinian evolution.

I am not even sure I buy the much more limited conclusion of the review that healthful lifestyles are transmitted to children. This seems like an awful lot of extrapolation from limited research. It is plausible and consistent with existing research that being obese or overeating during pregnancy may cause and epigenetic signal of abundance, leading to children with a greater tendency to put on weight. What is not clear is if this is clinically relevant in humans. I also would not generalize this to “healthful lifestyle.”

Crouch goes way beyond overcalling epigenetics, and takes a left-turn into bizarro world:

It paves the way for a review of the work of French biologist Jean-Baptiste Lamarck, whose theory that an organism can pass to its offspring characteristics acquired during its lifetime was largely ignored after Darwin’s publication of On The Origin of Species in the mid-1800s, that work defining evolution as a process of incidental, random mutation between generations.

Wrong. It’s not Lamarck’s theory, but worse still epigenetics has absolutely nothing to do with the mechanisms of evolutionary change. Crouch does then pivot to the actual subject of the review, the effect of parental habits on the child. He quotes one of the authors, Sarah Robertson:

“People used to think that it didn’t matter because a child represented a new beginning, with a fresh start.

“The reality is, we can now say with great certainty the child doesn’t quite start from scratch. They already carry over a legacy of factors from their parents’ experiences that can shape development in the foetus and after birth.

“There is now biological evidence that memories of experiences in adults can be transferred through egg and sperm for the lifetime prospects of the child.

“If evolution has developed something like this it can give a child an edge to survive. This will rewrite long held views, that experiences can actually be transferred to offspring.”

You will notice that Robertson never says anything to support either the notion that epigenetics calls Darwinian evolution into question, nor the hyped interpretation that lifestyle transfers to children. Everything she says is compatible with a conservative interpretation of epigenetics. Crouch, however, uses them as if they support his ridiculous interpretation.

This is unfortunately common in bad science journalism. The reporter has their story in mind, usually some hyped interpretation not supported by the actual research they are reporting, or some tiny footnote they will exagerate as if it’s the main finding of the study. They then interview experts, not to find out what is really going on or to put the story into perspective, but to mine for quotes they can plug into their pre-existing narrative.

Then he does it again:

Prof Robertson stressed that genes remain the blueprint for a new baby, but said the work of both Darwin and Lamarck may need to be reconsidered.

“The genes are the blueprint and that won’t change,” Prof Robertson said. “But this is at another level, it is the decoration of the gene, the icing on the cake if you like, a gift to offspring that gives them another layer of information about survival.”

She says nothing about Darwin and Lamarck, just like the review says nothing abuot Darwin and Lamarck. Her statements that he uses to support his interpretation, in fact contradict it. Epigenetics is another level of information, which does not change the basic fact that genes are the “blueprints.’

As an aside, I do not like the “blueprint’ metaphor for genes. It’s misleading. Genes are not a blueprint. There is no representation of the final product in the genes. Genes, rather, are more of a recipe, a set of instructions that, if followed, result in the final organism. Yes this is nitpicking, but metaphors in science communication should strive to be conceptually accurate.

He finishes with the lifestyle theme again:

“Lifestyle changes by potential parents and improvements in the right direction, especially in the months leading up to conception, could have a lasting, positive benefit for the future of their child,” Prof Robertson said.

This, at least, is the proper focus of the paper. Here Crouch failed also. He never mentioned epigenetics, never gave this central concept to the review article any background, and he did not appear to interview other scientists to see if this one scientist, discussing her own paper, was in the mainstream or not.


Crouch’s reporting of this review article on epigenetics was an utter disaster. He displays everything that is wrong with bad science journalism.

Epigenetics is also not a new concept, as one might think from reading Crouch’s article. It is still relatively young, scientifically speaking, and there is likely scant public awareness of epigenetics. Reporting on this review was a good opportunity to introduce readers to the concept of epigenetics and to put it into some historical and scientific perspective.

Anyone reading Crouch’s piece who was not already familiar with evolutionary theory and epigentic would have come away without gaining any real insight, but rather completely misinformed.

My own take on epigenetics is that this is an interesting layer of complexity to the function of genes that allows for short term adjustments to current environmental conditions. Epigenetics have no significant impact on evolutionary theory, however.

Also, the concept that genes themselves are not destiny, but rather genetic expression interacts with the environment is nothing new. The term “epigenetics” was coined in 1939, and the modern sense of the word (mechanisms that affect the expression of genes) dates back to the mid 1970s.

Here’s Crouch’s headline: “Darwin’s theory of evolution challenged by University of Adelaide genetic memory research, published in journal Science.” (I know reporters don’t usually write their own headlines, but this is fairly based on Crouch’s reporting.)

Here’s the real headline: “Local scientists write a review article on the decades-old concept of epigenetics.”

83 responses so far