Wednesday, October 30, 2013

On Genetic Determinism

23andMe.com is the website of a business that does a genetic analysis from saliva samples submitted by its customers.  In particular, they report on health risks.  Following Bee, I had submitted a saliva sample to them. One new thing I learned from their technical reports was that for incidence of disease,  the very same gene variants had different sized effects in people of European, African or Asian ancestries. 

There could be a number of reasons for this - the interaction of these gene variants with other genes or environmental and developmental differences,  or epigenetic differences.   In any case, these effects throw sand into the gears of naive genetic determinism.

If you want examples from 23andMe.com you are going to have to subscribe to them.  Here is an example from Wiki:
The same gene variant, or group of gene variants, may produce different effects in different populations depending on differences in the gene variants, or groups of gene variants, they interact with. One example is the rate of progression to AIDS and death in HIV–infected patients. In Caucasians and Hispanics, HHC haplotypes were associated with disease retardation, particularly a delayed progression to death. In contrast, for African Americans, possession of HHC haplotypes was associated with disease acceleration.
We might get "lucky" and human intelligence might be universally determined by a handful of genes in which we can determine the "good" and "not-so-good" variants.  That opens the door to genetic engineering of humans for improved intelligence.  Or we might discover that Mendel discovered laws of heredity by happenstance, having picked traits in his pea plants on which simple genetic determinism does work; but that that is largely the exception, not the rule for features like intelligence.   We might get "unlucky" and find out that human intelligence is governed by complex interactions between genes with any particular variant of a particular gene being advantageous or deleterious depending on the entire set of other genes; or we might get "really unlucky" and find that intelligence depends not just on genes but the entire history of the individual (or even on the history of the individual and its previous generations - that pesky epigenetic thing again).

I think how it is going to turn out is very much an open question.  I think only ignorance or ideological predilections lead one to assert that it is definitely going to turn out one way or the other.

Further, the possibility of genetic engineering is not going to raise problems that we don't already have in some form.  In India and in China, already the XX chromosome pair is selectively engineered against despite laws against it,  doctors being required not to share ultrasound images of the fetus with their patients and so on.  In the state of Haryana in India, the gender ratio is at around 877 women per 1000 men.   The dire effects on society are already being felt.  Unfortunately, the modern ease of determining the sex of the fetus cannot be regulated away.  One can only hope that other forms of genetic engineering are less accessible and thus more susceptible to regulation.

Since tens of millions of women were never even born, some activists term it a genocide.

That is why I find the following exceedingly simplistic, especially coming in a post that makes the accusation of large-scale willful blindness.
Suppose we do find that higher IQ is closely associated with certain genotypes. What then? The most likely thing in the future is not some kind of genocidal murder program, but rational therapy, based on genetic medicine. Echoes of Brave New World? Yes, but it's coming, and we ought to prepare to deal with it.