The Diabetes Gene

Research published in the journal Genetics reports on “the gene responsible for traits involved with diabetes.” That news might lead you to believe that we can now predict who will need treatment for diabetes, how to treat those people, and maybe even how to prevent those people from developing diabetes in the first place.

Nothing could be further from the truth.

This research is just the most recent to emerge from the Human Genome Project and the technologies it spawned. The promise made by Project advocates was that tremendous progress would be made in preventing and curing illness, especially chronic and costly diseases such as diabetes, cancer, Alzheimer’s disease, heart disease, and so on.

Nothing like that has happened. Instead, lavish sums continue to be spent on research such as this diabetes study that identified associations between sections of DNA and susceptibility to poor blood sugar and insulin control.

I do not have a quarrel with genomic research. It helps us better understand how living things live. What I object to is the false promise of genomic research and the resulting failure to fund other research that will really help prevent and cure those chronic and costly diseases.

The false promise takes advantage of what we civilians think is going on when genomic discoveries are made and what is actually going on in the science. There’s two parts to the falsehood.

The first is what we think a gene does. The sense we have is that a gene determines health and illness. We think there is a gene for diabetes, a biological entity that causes the body to do a poor job of metabolizing glucose and producing insulin. It’s quite similar to the 19th Century idea that who and what you are is inherited from your ancestors.

On the other hand, genomic science is working on the biochemistry of DNA and its association with the likelihood that someone has diabetes or will get it. For example, the research I mentioned associates a stretch of DNA possessed by humans, mice, and rats with how the metabolism of each responds to glucose: those animals with a specific variation of the DNA sequence are more likely to overproduce blood sugar and underproduce insulin.

That isn’t the same as the gene that causes diabetes. DNA is the material of the gene in the same way that neurons are the material of the mind. Using the word “gene” to refer to a DNA sequence is what’s called in philosophy a category mistake, just as using the word “mind” to refer to a clump of neurons is a category mistake.

You might think, “So what? They can call it whatever they want, just as long as I don’t get sick.”

You might also want to think about the time that’s wasting while researchers “identify genes and variants of interest’ with little effect on preventing, treating, or curing the disease because they aren’t getting at the cause.

The second false promise is the resulting sleight-of-hand in which a DNA sequence is made to appear as the cause of a disease. On closer inspection, it’s quite clear that the link between DNA and diabetes trait is not causal but statistical. In other words, if you have the variant of this new diabetes DNA sequence, it only means you’re more likely to have poor glucose and insulin control compared to someone with another variant. It doesn’t mean you will have poor glucose and insulin control. The variant does not cause the trait.

You might think, “So what? At least I know I’m at risk and can take appropriate action.”


In statistics, there’s a pair of concepts called explained and unexplained variation. The DNA researchers looked at human, mouse, and rat DNA and found that poor glucose and insulin control tended to occur in those with the variant DNA—but not in all those with the variant. In other words, the presence of the DNA variant explains some but not all of the variation in glucose and insulin control. In fact, it only explains a small amount. And note that the word here is “explain” not “cause.”

What explains the rest? The researchers do not know—in part because they do not ask. That’s left to others. The most interesting people asking are working outside genomics.

Epigenetics is one such field: it looks at environmental factors such as nutrition, nurturance, and other exposures as what turns on and turns off chunks of DNA—in some cases, affecting future generations.

Public health and preventive medicine are two other fields. Yet another is environmental or more broadly ecological health—that is, how the natural environment is affected by and affects human health.

As far as human health goes, genetic research since the completion of the Human Genome Project has been a bust. The best that can be said is that some discoveries can inform us about our susceptibility.

However, I believe its most important result is to make it abundantly clear that the principle cause of illness is environmental. Eliminate the environmental exposure that provokes a troublesome chunk of DNA and susceptibility becomes irrelevant.