Friday, December 7, 2018

How To Migrate To America Without Leaving Much Of A Genetic Trace


The only genetic trace of any population other than the Founding Population with a small effective size arriving in the Americas at any time prior to the ca. 3500 BCE to 3000 BCE time frame when the Paleo-Eskimo ancestors of the Na-Dene peoples arrive is a smidgen of Paleo-Asian ancestry (on the order of 5% to 10%) in a couple of groups of tribes in the Amazon jungle near the base of the mountains to their west and north, who number in the aggregate, closer to 10,000 people than 100,000 people.

The bulk of the Founding Population migration from Beringia to the rest of North America and South America at which point the Founding Population starts to grow exponentially until it reaches the full carrying capacity of those two continents with the technologies available to them at the time, starts around 16,000 years ago, although it is possible that some members of the Founding Population jumped the gun in modest numbers whose population growth didn't explode the way that it did around 16,000 years ago.

What pre-Na-Dene migrations to the Americas could have happen that wouldn't leave a genetic trace that has been detected so far (considering primarily genetic evidence at this point)?

Any scenario in which there is a genetically distinct population of modern humans in the Americas other than the Founding population at any time more than 6000 years ago or so, must fit into one of the scenarios below to have escaped detection through population genetics at this point.

In General

Since at this point DNA testing is of statistical and opportunity samples, rather than being comprehensive, anything that keeps that target DNA from dispersing into a large generation population and experiencing some approximation of panmixia rather than staying in clumps in a small discrete population that is small enough, or hard enough to recognize as distinct, can allow it to escape inclusion in any samples examined so far. So, while low absolute levels of the target DNA in the macro-population is important, strong clustering as opposed to dispersal of the target DNA also matters a great deal.

Among other things, it is critical to keep the undetected or exceedingly rare population out of any exponentially growing founding population.

Bottleneck conditions during adverse environmental circumstances or after wars are tricky. Outsiders who arrive right before a bottleneck can easily have their genetic trace lost to random drift as the population shrinks. But, if any of the outsider DNA manages to stay in the gene pool once the population rebounds after a bottleneck, that outsider DNA will be caught in a founder's effect and become widespread and common in the successor population.

1. Into the Americas That Failed.

It is possible that modern humans or more archaic hominins arrived in the Americas before the Founding population did but went almost entirely extinct before they arrived for some reason. Perhaps some relict population survived in some place that had effective boundaries to prevent their expansion beyond a small population out of their refugium. Perhaps that relict population experienced genocide at the hands of the Founding population upon first contact in most locations so that only a few members of the relict population introgress into the Founding population gene pool and those genes are lost due to genetic drift in the first few generations. This scenario is especially possible if the encounter with the relict population happens relatively close to the time that the Founding Population has expanded to its Malthusian limit. Otherwise, even a minor introgression of genes into the Founding Population during a period of rapid population growth for the Founding Population would be magnified by a Founder's effect and it would be almost impossible to prevent the introgressed genes from becoming ubiquitous in a moderately sized region of the Americas within a few generations of the initial introgression.

Examples: Modern humans has expanded beyond Africa by ca. 125,000 years ago, and has reached India before 75,000 years ago, but did not experience rapid expansion until about 75,000 to  50,000 years ago.

Madagascar had a small modern human population that did not rapidly expand in population or thrive economically for at least significant portions of the time period five hundred years or so before the Austronesians arrived. But, when Austronesians arrived, all trace of them vanished if they had not gone extinct themselves before the Austronesians arrived.

In the Americas, Leif Erikson's attempt to colonize North America from Iceland ca. 1000 CE, and the Roanoke colony are examples of failed attempts at colonization that left no subsequent genetic trace.

2. Migrations Of People Who Are Late To The Party And Not Conquerors.

Once the Founding population expands to its Malthusian limit and populations stabilize or even experience occasional bottlenecks in some location, a much larger introgression can go undetected for a long time, because the genetic effect of the introgression isn't amplified into a Founder effect, and the likelihood the same of the introgressed genes are lost to genetic drift over time is much greater, especially if the introgressing population has a fertility rate or population genetic fitness that is not larger than that of the existing population. It also helps if the population into which the introgression takes place is geographically immobile, perhaps because of geographic boundaries or because neighboring regions are populated with people who don't welcome newcomers.

Examples: Gypsies in Europe. Jews and Catholics in the Deep South of the United States or Appalachia.

3. Closely Related Populations.

The more similar a population other than the Founding population is to the Founding population genetically (which usually implies a most recent common ancestor at not too great of a time depth), the harder it is to distinguish through genetic data from a single wave Founding population, so long as the additional population also has a small effective population size. This is because if they had a large effective population size they would make the Founding population effective population size look larger, and we know that the apparent size of the Founding population, robustly through many kinds of measures, is very small, probably the smallest of any continent sized population apart from the Aborigines of Australia and indeed probably even smaller than in Australia.

Examples: Recent English immigrants to New Zealand are genetically indistinguishable from the earliest English colonists of New Zealand.

There were probably at least two waves of Indo-Aryan migration into India separated by several centuries, one of which reached the whole sub-continent, and the other of which only extended to part of the sub-continent (generally in the north). But, the two waves of migration in northern India are very hard to distinguish from a single Indo-Aryan migration wave.

4. Isolated Outsiders.

The non-Founding population is small in absolute number, is strongly endogamous for the time period until the Founding population (at least locally) is at the tail end of the logistic population growth curve, and the non-Founding population has a significantly lower population growth rate than the Founding population for some reason.

The endogamy doesn't necessarily have to be primarily a social matter, although it can be. Geographic barriers such as long distances of water that must be traversed or high mountains or deserts or jungles can also powerfully enforce endogamy.

The most extreme example of this would be a single person (think Marco Polo) or a family (think Swiss Family Robinson) that comes to the Americas and doesn't have any children after they arrive who survive to have children of their own.

Example: Jewish communities in Africa and India. Religious minority populations in Iran. North African sailors in the British Isles. There are quite a few examples of very strong (although rarely absolutely and entirely perfectly maintained) endogamy for long periods of time (a couple thousand years or so), so as jati endogamy for roughly the last two thousand years in India.

Different Levels Of Impact

Historically, replacement of Y-DNA in a population is most common. Replacement of mtDNA in a population is much more rare but can happen in isolated circumstances. Complete autosomal DNA replacement is much more rare than complete replacement of either kind of uniparental DNA.

This probably happened to some extent in the founding population of the Americas which has much more mtDNA diversity (even though that itself is very modest) than it does Y-DNA diversity.

Similarly, while all modern humans out of Africa have a small single digit percentage of Neanderthal ancestry, no Neanderthal Y-DNA or mtDNA survives in modern human populations. The same pattern is observed in the smaller subset of people who have Denisovan ancestry.


Guy said...

Hi Andrew,

I particularly like the 2nd para which is one sentence.

1. Moving into a virgin territory and not expanding? That take very special pleading. On an island you might not be able to find a nice place to live, but on two continents with no competitors?

2. Yep, Sinbad took them over and left them.

3. But the Paleo-Asian is distinguishable and distinctive.

4. For ten thousand years... I rather believe hominids that also mixed into the Paleo-Asian root stock.


andrew said...

Re 4. Not for ten thousand years, it is only necessary to be "strongly endogamous for the time period until the Founding population (at least locally) is at the tail end of the logistic population growth curve". In a stable population with small N of the outsider group, only weak endogamy is necessary because genetic drift filters the gene pool much more efficiently then.