The Soledad O'Brien Effect

April 14, 2006 4:50 PM

Say that interethnic breeding has created a homogenous-looking mixed-race human species in the far-flung future. What kind of effect would this have on biodiversity?

Occasionally, during discussions of race relations and interracial marriage, someone will bring up such a future in conversation and call it an ideal solution: we've interbred so much that there's no longer any real distinction between races, visual or otherwise, and thus whatever conflicts we have related to visible minorities will simply crumble to dust. Then someone else will pipe up about how such a human race would become more vulnerable to disease or extinction because it no longer has the genetic diversity inherent in multiple races.

Ignoring the political aspects of the light brown future (because honestly that's a different subject altogether), does the genetic diversity argument make sense? Would we really be shooting ourselves in the foot if we blurred the divisions between ethnic groups to the point where they no longer existed? Or is genetic diversity independent from the notion of race/ethnicity?

posted by chrominance to Science & Nature (22 answers total)

Check out this link.
posted by billysumday at 4:56 PM on April 14, 2006

I doubt it. There's already six billion people on this planet and it's just going to keep going up. There's no way that everyone would be genetically homogeneous enough to cancel out "biodiversity".
posted by borkingchikapa at 4:57 PM on April 14, 2006

Generally, biodiversity is used to discuss variety at the species level and above; genetic variety is used to discuss the variety of genes and amount of heterozygosity, which is what your question is getting at.
posted by docgonzo at 5:18 PM on April 14, 2006

Best answer: Your question gets the problem backwards. While interracial breeding spreads genes more evenly across the population, any two individuals are in fact more likely to have highly divergent genomes than in a racially segregated population.

It helps to remember that genes don't actually "mix" -- even though they sometimes appear to when it comes to skin color.
posted by Ptrin at 5:37 PM on April 14, 2006

Best answer: The real answer is, almost none at all.

The differences that contribute to racial differences have been established so recently as to be insignificant. There were three or four recent migrations that led to the races that you see now. The groups that formed the European or Asian migrations are simply bottlenecked populations that expanded to fit their new homelands, which is to say that those populations are the ones lacking in genetic diversity because they were effectively founded by tiny tiny populations and fairly recently as well. The most diversity can actually be found among the old (African) populations which have had a long time to acquire various adaptations and aren't all homogenous.

The effective population size of the human race is only 10,000 individuals, just because of all these bottlenecks and die-offs that have occurred. Effective population size refers to the equivalent randomly-breeding (ideal) population that has the same allele frequencies (genetic diversity in your language) as our real-world, non-randomly, highly bottle-necked population.

To re-unite the various wings of human migration would not substantively change the amount of genetic diversity present in the human popluation. Quite simply, the races have not diverged for a long enough time to create a significant pool of differences. Our mitochondrial Eve (theoretical woman to which all humans can be though to descend from (probably from Africa)) is estimated to have lived 150,000 years ago, which is the blink of an eye, in evolutionary time.

The balance of different allele frequencies (gene variants) would be the same no matter if we were separately white, black, yellow, and brown, orl a light mocha a couple thousand years from now.
posted by Mercaptan at 5:37 PM on April 14, 2006

Pretty much everyone else's consensus here, too.

There are a *lot* of recessive alleles. Red-headedness will still be around, but probably more-rare.

However, if a smaller group of individuals went and colonized another planet, there could still be a founder effect where red-headedness might eventually become predominant (for example, if there were positive selection pressures [perhaps sexual attractiveness] for being a redhead).

Feel free to subsitute redheadedness for any number of recessive (or complex) traits, both positive and likely negative (negative traits may persist because onset of selective disadvantage occurs after peak procreative age) too, especially if being a heterogygous carrier is more beneficial than being homozygous positive (even if being homogygous negative is a very very bad thing).

Also, it's not too far fetched to imagine that humans will still care for each other - the increased medical intervention in childhood diseases is increasingly affecting genetic counseling in partners who wish to procreate. I don't have the numbers in front of me, but I'm guessing that there are more type I diabetics (completely genetic) now than there were 200 years ago (say, waaay before the discovery of either bovine/porcine or recombinant insulin). Medicine may help "negative" traits persist in a dynamic population.
posted by PurplePorpoise at 5:39 PM on April 14, 2006

Or is genetic diversity independent from the notion of race/ethnicity?
Yes.

Any offspring by sexual reproduction is genetically distinct from its parents. It's a brand new combination of genes with traits that are visible and invisible. You can't think of genes as paint - it's not a homogenous mixture from two sources. Rather it's more like drawing a combination of lottery balls from two baskets at random.
posted by junesix at 5:46 PM on April 14, 2006

Best answer: "Race" as perceived is in part a function of combinations of genes which tend to occur together, for reasons of history and geographical separation. So "Northern European" means a significant probability of blue eyes, and/or light colored hair, almost universal fair skin, invariant round eyes, large body size, larger breasts on women, and hair which is straight or wavy. "Sub-Saharan African" means a higher probability of dark skin, virtually universal round brown eyes, curly hair, large size, larger breasts on women. "Japanese" means small size, smaller breasts on women, sallow skin (or light tan), invariant brown eyes and almost universal black hair which is straight, neotenized physiques. (There are exceptions to most of those; I'm trying to describe the norm, not the range.)

Given the future genetically-homogenous race you describe (and it would likely take millenia) it would have no important effect on the total amount of genetic diversity in the human race. All it would mean is that genes which now tend to occur together no longer would.

The genes themselves would not be destroyed; it doesn't work that way. So you'll still have the genes for blue eyes and light hair running around. There will still be genes for straight hair, wavy hair, and curly hair. There will still be genes for a variety of different skin tones. There will still be genes for round eyes and for slanted eyes. But the correlations between those genes which now exist would be broken.

You'll never get a situation where everyone looks the same e.g. everyone's hair is black or brown, everyone's eyes are brown and round, everyone's skin is medium tinted. Rather, you'll start getting combinations which we rarely see now absent cosmetics, e.g. sallow skin with red hair and slanted blue eyes, neotenized features on big bodies.

Probably the single most important rule in genetics is that genes in heterozygous individuals don't mix. They may both express or may not, but they remain separate and each can be inherited separately. Recessive genes can be around for a dozen generations without expression, but when they double they're just as they were a dozen generations before.
posted by Steven C. Den Beste at 5:51 PM on April 14, 2006

Post script: My dad's hair was black. My mom's hair was dark brown. They had three kids and all of us were redheads. That's how it works with genetics and recessive genes.
posted by Steven C. Den Beste at 5:55 PM on April 14, 2006

if you're genetically divergent enough from a second cousin to breed without problems, and if (say) a third of the world's population is white, another brown, and another yellow, wouldn't it make sense given your postulate that it's more harmful to breed within your race since it already gives less chance for diversity given a smaller population than this "single race"?

that's my layman's way of looking at things.
posted by kcm at 6:06 PM on April 14, 2006

Race has no biological meaning. Given that, your question doesn't make sense. If you are just asking if widespread interbreeding will affect genetic diversity, then (I think) the answer is a qualified "no". (A widely interbred population will begin to show selection pressures on various genes that did not exist when those genes were in an isolated population.)
posted by Ethereal Bligh at 6:10 PM on April 14, 2006

While EB is right in that race is not a true biological concept, it is valid in as much as it is a proxy for the genetic structuring of the human population. It is well-known that certain populations have higher frequency of certain alleles which predispose to certain conditions, ie sickle-cell or CCR5-delta32.
posted by docgonzo at 7:08 PM on April 14, 2006

Response by poster: Okay, so let's see if I've got this straight: the more mixed a population is, the less likely a certain combination of genes is likely to dominate in that population; therefore that population will be more resistant to disease vectors that somehow target that particular combination of genes.

If we were to consider our "population" to be all of humanity, we'd see no real difference in our ability to survive genetically introduced susceptibilities to disease since there'd be just as many vulnerable genes (and perhaps fewer vulnerable gene combos, if genetic diseases work that way—I'm such a noob when it comes to this stuff), and if we take a subset of that population—say, everyone living in Japan—they'd actually be more resilient given a light mocha future, because whatever susceptibilities might be present now have diffused into all of humanity. Does this sound right?
posted by chrominance at 9:26 PM on April 14, 2006

Response by poster: (oh, and it's also good to know what the consensus is on race as a biological versus social classification, special thanks to billysumday for the link.)
posted by chrominance at 9:28 PM on April 14, 2006

For sound evolutionary reasons, the vast majority of negative genes are recessive. The very few exceptions almost all have delayed effects e.g. Huntington's disease, or are extremely rare e.g. retinitis pigmentosa. (There are also a few which are codominant, such as sickle cell disease and cystic fibrosis, about which more below.)

Recessive genetic diseases like Tay Sachs only manifest when doubled. A person who is heterozygous (a "carrier") has no negative symptoms and often there's no indication at all in the phenotype that such a recessive gene is present.

The chance of doubling is a function of the prevalance of such recessive genes in the population. If 1% of the population carries the gene, then 1 baby out of 10,000 will hit the jackpot. If 5% of the population carries the gene, then it's 1 baby out of 400.

Tay Sachs is a genetic disease which historically was present in Ashkenazi Jews, and because for political and cultural reasons they tended to marry one another for the most part, and because the percentage of carriers was significant, Tay Sachs disease had shown up quite a lot amongs them.

There are other diseases like that in other racial groups. But if a man from one group who is a carrier of one bad recessive marries a woman from a different group who is a carrier of a different recessive gene, then it's possible for their offspring to be carriers for both but it isn't possible for them to be afflicted with either disease. (Geneticists refer to this as "hybrid vigor", and it applies to all kinds of things. It's the reason why mutts are healthier than AKC dogs, for instance.)

If in-group breeding preferences were largely abolished and there was lots and lots of genetic mixing world wide, those negative genes wouldn't disappear. But their prevalence would decline substantially, and the chance of them doubling would drop considerably. Not to zero, but by quite a lot.

Which nonetheless doesn't necessarily mean the number of cases per year would decline. The chance per baby would drop, but you'd have more babies at risk. Sigh. It would vary from allele to allele.

As to the question of whether such grand mixing would make us less vulnerable to world-wide plagues, probably not by much. You gain some things with one hand but lose others with the other hand.

Some forms of genetic resistance to disease only manifest in the phenotype if they're homozygous. The general trend genetically would be to make everyone heterozygous. So you'd have a decrease in the number of people getting genetic diseases, but also a decrease in the number of people who were homozygous with a characteristic that protected them against certain kinds of diseases.

On the other hand, certain genetic traits protect against disease only if they are heterozygous. That's the case with both sickle cell disease and with cystic fibrosis.

A person who doubles the sickle cell gene gets "sickle cell anemia" and without modern medical treatment rarely lives to adulthood. But that gene is codominant; which means that even if you're a carrier it expresses. A person who is heterozygous is different from one who is homozygous with the sickle gene, and is also different from someone who is homozygous with the normal gene. And the reason that's good is that a person who has one (but not two) copies of the sickle gene is not affected by sickle cell disease but is highly resistant to malaria.

Cystic Fibrosis is the same way. A person who doubles that gene has really bad lung problems and in the old days usually died as a child. But like the sickle gene, the CF gene is codominant, and a carrier of the CF gene has no lung problems, but turns out to be less likely to die from serious diarrhea, such as that caused by cholera or amoebic dysentery. That's because they lose less potassium and other electrolytes in the fluid flow.

So in your hypothesized future, we get a lower rate of genetic diseases (e.g. Tay Sachs), a lower rate of resistance to some diseases (because they require doubled genes) and a higher rate of resistance to some diseases (because they require non-doubled codominants).

When it comes to genetics, nothing is simple or straightforward.
posted by Steven C. Den Beste at 10:00 PM on April 14, 2006

By the way, some people think that the reason the CF gene is so prevalent among northern Europeans is because carriers of the CF gene were more likely to survive the Black Death. No one knows that for sure, but it's not impossible.

There's no question that the reason why the Sickle gene is so common among sub-Saharan Africans is because of the threat of malaria. Where malaria is common and dangerous, the Sickle gene saves more lives than it destroys, and for evolution that's enough.
posted by Steven C. Den Beste at 10:07 PM on April 14, 2006

"it is valid in as much as it is a proxy for the genetic structuring of the human population."

No, that's exactly the way in which it's not valid. There is no reliable correlation between race and genome from a pool of all human beings. In many cases, however, if you limit the pool—for example, north Americans—then you will find a correlation. But across the entire human population, the characteristics we associate with "race" are not reliable indicators of relatedness.

Common sense seems to indicate that this isn't the case and that there's a close correlation between race and genome. The reason that this isn't the case, however, is that the characteristics we have selected to bace racial distinction upon are mostly superficial characteristics. The trait about which this is most true, and notoriously true, is skin color. Skin color plays a huge role in making racial distinctions between different populations; but in fact it is not a reliable indicator of genetic relatedness because a combination of genes control skin pigmentation, quanitification of that pigmentation is ambiguous and lies on an continuum, and two independent populations may have completely different genetic antecedents that result in the phenotypal distinctions around which the social concept of "race" is determined. There is no single, distinct gene that accounts entirely for skin pigmentation. And even if there were, two independt populations could have that mutation arise independently, and thus it wouldn't be an indicator of relatedness. (It might arguably indicate a certain shared quality. But in the real world skin pigmentation is not controlled by a single gene and so this is not true, either.)

The degree of corrleation of racial distinctions as they are socially defined and the genetics of various populations will vary if we break these racial distinctions down into more discrete characteristic markers, such as skin color. Another marker might be the fold of the eyes, or something else. While skin color is a very, very unreliable indicator of relatedness, eye shape might be somewhat more reliable. Even so, the ultimate problem here is that our social construction of "race" is built around a cluster of percieved physical differentiations that range biologically from genetically insignificant to moderately significant. The end result is that you could probably very selectively choose a few characteristics that are related to our intuitive concept of race that do, in face, moderately correspond to a genetic relatedness. However, that is essentially stacking the deck. If you were to use a simple list of the few main physical characteristics that are widely agreed upon to differentiate "race" you would not be able to find even a moderate genetic correlation to relatedness in a pool of the entire world's population. In the cases where you are carefully selective of the characteristics and they are reliably correlative to relatedeness, you've merely created a tautology: you find a population that is closely related and find a physical characteristic among many commonly associated with "race" that correlates strongly to relatedness. You're going backward, that's not science.

It's hard for people to understand that racial distinctions do not correlate to genetic distinctions. We know that dark skinned parents have dark skinned offspring, for example. But within certain populations, the same thing can be said of hair color: light hair colored people have offspring that are also lightly colored but we don't, in those cases, assume that every light hair colored person in the world is more related to each other than they are to dark hair colored people. It is a superficial trait that doesn't represent a deeper truth. We think it does in the case of race and skin color, and so there is folk widsom about all black people with regard to, for example, their bone density (as an explanation for the paucity of black swimmers) or some other very intrinsic quality. In fact, however, while we can imagine that skin color signifies some deep qualitiative genetic differentiation between population, the truth of the matter is that it doesn't. The biological, genetic status of "race" is false because it happens to be false, not because it necessarily must be false. This is why it's such a pernicious concept—it has a certain social utility, it's built around some basically correct ideas about genetics, and it could be true...even though as it happens, it isn't.

A very clear cut example of how race is assumed to be biologically meaningful but in fact is not, is the example that docgonzo provides us—sickle-cell anemia. He says that this mutation correlates to race, but in fact it does not. Most black populations, in actual fact, do not have the sickle cell mutation and so it's not a reliable indicator of blackness, nor blackness a reliable indicator of it. Only a small portion of the indigenous African population has this mutation, the rest do not. The populations that have it are populations where malaria has been endemic for very long periods of time. This is western African populations. And, as it happens, the majority of black Africans taken as slaves to America came from that same west African population. The result of this is that within the American black population, there is a strong correlation between the sickle-cell gene and genetic relatedness. And this is true for a very simple reason: the black American population largely shares the same few common ancestors from that part of west Africa. However, that west African population is very far apart, genetically speaking, from some other African populations and, as it happens, some black African populations are more closely related to white northern European populations than they are to other black Africans.

So this is another reason "race" is a pernicious concept in the US: within the limited pool of north Americans, black Americans really are largely related to each other. This allows certain forensic techniques and other biological generalizations about race to actually be usefull. But make no mistake: if we were to replace the black American population with people randomly picked from all black populations worldwide, suddenly all those forensic techniques and the like would no longer workd. Because, it's worth repeating again and again, skin color does not correlate with genetic relatedness.
posted by Ethereal Bligh at 11:12 PM on April 14, 2006

I blew a calculation: If 1% of the population carries the gene, then 1 baby out of 10,000 will hit the jackpot. If 5% of the population carries the gene, then it's 1 baby out of 400.

Actually, if 1% of the population carries the gene, 1 baby out of 40,000 hits the jackpot. If 5% of the population carries the gene, it's one baby out of 1600.

It turns out that about 4% of people of northern European extraction carry the Cystic Fibrosis gene, which means it expresses as a double recessive in one out of 2500 babies born to such parents.

The Sickle gene is even more prevalent among some groups of people of African descent.
posted by Steven C. Den Beste at 11:52 PM on April 14, 2006

There was a really interesting series on NOVA --"In Search of Human Origins", I think-- where, using genetic markers, they traced back to what they believe to be the founding population for humans. It was a tribe in Ethiopia that still exists today, so they went to speak with them. The crazy thing was, they observed that the people looked an awful lot like an amalgamation of what we associate with separate racial traits: brown skin, light brown kinky hair, freckles, high cheekbones. They looked like a blend, and you could see where some of these traits might have gained in frequency in isolated populations.
posted by Gamblor at 10:25 AM on April 15, 2006

As some other people have pointed out, the notion of "race" as a biological construct falls short when you realize there is more genetic diversity within Africa than across the remaining continents combined.

But I also think this ignores some other problems. First, cultural and ethnic homophily are still strong drivers of marriage and reproduction. Second, Homo sapiens certainly has not stopped evolving, so you still might find genetic drift leading to different gene frequencies in relatively isolated population groups.
posted by KirkJobSluder at 5:53 PM on April 15, 2006

Although that was an excellent post, I offer a shorter EB: what most people think of as 'race' is just a spurious manifestation of the human brain's predilection towards pattern recognition.
posted by stavrosthewonderchicken at 6:52 PM on April 16, 2006

I would like to hazard a guess that chances are; given a significant enough period of time for science and morals to advance; that people will likely choose their children's physical characteristics from a directory. At this point you'll see diversification as well as "ideals" such as Aryan dickheads breeding their children as blonde haired, blue eyed ubermenschen and goth parents with their perfect little minigoths.

What this will do to society is extremely interesting to me. Assuming we still have not gotten past the stupid racism thing by this stage expect to see an awful lot of enclaves pop up (a la Diamond Age) except not just social but physical characteristics defining them.
posted by longbaugh at 4:52 AM on April 17, 2006