genetics are not as predictive as we might think -- 8/24/15

Today's selection -- from Evolving Ourselves by Juan Enriquez and Steve Gullans. Genes are less predictive of behavior and disease than is generally assumed:

"Much of the general public ... has ended up with the mistaken impression that almost every imaginable disease has a clear and readable genetic basis. But very few diseases operate like a light switch or correlate in a one-to-one way whereby if you have gene X then you get disease Y. Time and again, studies fail to find the genes or genetic variants responsible for a majority of the cases of heritable diseases or traits, including for Type 2 diabetes (only predictive in 6 percent of a population), good cholesterol (5 percent), early myocardial infarction (3 percent), and familial breast cancer (10 percent). In every case, a few genes are identified as causal, but in only a small subset of patients. This includes the recent sensational headlines that 'specific gene variants' predict whether taxes deter cigarette smoking in one group versus another. You have to read the article, and the footnotes, before you realize this is true, but only for 1 to 2 percent of the population.

"So when you read yet another story headlined 'Scientists Find a Gene for X,' read the endnotes, qualifications, and caveats carefully. They may have found a statistically significant correlation, but it's usually only relevant to small subsets of a population. For the many complex diseases that run in families, including cancer, hypertension, neurodegeneration, autoimmunity, diabetes, obesity, schizophrenia, and depression, the majority of supposedly causal genes are missing. Typically, in the strongest correlations, scientists uncover genetic differences that account for only 5 to 15 percent of the cases. This nasty, large, dark secret among the lab-coated is known as 'missing heredity.'

"Not finding specific, causal, predictive genes for inherited traits, diseases, and conditions remains far more common than finding the specific genetic signatures that allow one to say, yes, a person has this, will develop that, will grow to be that. But geneticists, being very smart people who often trained in mathematics and physics before switching into life code, focus on results and success, and rarely write commentaries about why they can't find the missing genes.'

The genome is folded into a complex
but highly organised ball

"The occasional 'missing gene' commentaries fall into a category all their own; these are among the most difficult and densest papers you will ever attempt to read. ... When you finally drag yourself to the conclusion of anyone of these papers, you basically read the same four explanations over and over again: (1) 'It's complicated and related to many, many interacting genes'; (2) 'We need more data (and certainly more grant money); (3) We will get the answers once we sequence the genomes of millions of people'; (4) 'The genes aren't really missing, they are just hidden and our new strategy will find them.' Or, if they truly found nothing, they might just adopt a serious demeanor, look over their glasses, harrumph, raise their eyebrows, and expectorate an old chestnut; 'It's a combination of nature-nurture ... '

"Despite billions of dollars spent, most labs suffer from constant cases of gene interruptus, leading everyone to focus on and endlessly cite the few genes and codes that do clearly correlate, leaving a deafening silence on the majority of conditions, diseases, and traits that do not have clear, mathematically probable, peer-publishable results. Grant after grant gets spent on results that basically say, 'My lab rats ate my genes.'

"Intelligence is a good example of missing heredity; a substantial part, perhaps 50 to 85 percent, is assumed to be inherited? But try as they might, gene scientists just cannot find 'the gene' or genes associated with this minor and non-obvious human trait. In a forced march, reminiscent of the early sieges of Everest, the Beijing Genomics Institute launched a controversial study of a mere 126,559 individuals in an attempt to find intelligence's missing heredity. After analyzing tens of thousands of small genetic differences that existed within this group, and looking for correlations, the study showed very minor effects from three variants within a few genes. How can it be that we cannot find any evidence of such a supposedly predictable genetic predisposition?

"Most human diseases, behaviors, and traits involve a combination of the genes we got from our parents as well as events we experience in our everyday lives, especially in utero and as children. You might even throw some microbes, or the lack thereof, into the predictive equations. Many of the conditions that seem to run in families, such as cancer, depression, intelligence, asthma, athletic prowess, height, addiction, happiness, autism, hypertension, musical talent, body weight, childhood aggression, longevity, altruism, heart disease, and schizophrenia, are part heredity and part environment."



Juan Enriquez and Steve Gullans


Evolving Ourselves: How Unnatural Selection and Nonrandom Mutation are Changing Life on Earth


Penguin Group


Copyright 2015 by Juan Enriquez and Steven Gullans


barns and noble booksellers
Support Independent Bookstores - Visit

All delanceyplace profits are donated to charity and support children’s literacy projects.


Sign in or create an account to comment

<< prev - comments page 1 of 1 - next >>