Tuesday, July 20, 2010

Evolving Humans

Nicholas Wade is often considered to be one of the best science journalists. He writes for The New York Times. His latest article is: Adventures in Very Recent Evolution. Here's one paragraph.
Many have assumed that humans ceased to evolve in the distant past, perhaps when people first learned to protect themselves against cold, famine and other harsh agents of natural selection. But in the last few years, biologists peering into the human genome sequences now available from around the world have found increasing evidence of natural selection at work in the last few thousand years, leading many to assume that human evolution is still in progress.
Later on in the article, Wade seems to be aware of the other mechanism of evolution but here he equates "evolution" with "natural selection." What do we have to do in order to educate science journalists? [Have Humans Stopped Evolving?] [Did biologists really think that human evolution stopped?]

Anyone who assumed that "humans ceased to evolve in the distant past" simply doesn't understand evolution. You can't stop evolution.

The main thrust of the article is whether natural selection is having a significant impact on our genetic makeup. There are many biologists who support the idea that more than 10% of our genes (alleles) are under selection and most of these biologists think that evolution by natural selection may even have sped up in the past 10,000 years.

John Hawks is a proponent of recent rapid human evolution by natural selection1 and, as expected, he has a post discussing The New York Times article [Recent selection, the new paradigm ]. I'm still pretty skeptical of those studies that claim to detect selection by analyzing genomes. I find the lack of agreement between different studies much more troubling than John does.

Recent talks and posters at the Society for Molecular Biology and Evolution meeting (SMBE 2010) highlighted some of the problems. Some emphasized the large number of false positives2 in published studies and questioned the accuracy of the algorithms. Others pointed out that biased gene conversion at recombination hotspots may be much more frequent that we assumed and this gives the appearance of selective sweeps when, in fact, the alleles being enriched may be neutral or even detrimental.

Since john hawks weblog doesn't have a comments section I thought you readers might like to discuss it here.

Photo Credit: The Future of Human Evolution [Aaron Avivi].

1. See Examples of Accelerated Human Evolution.

2. See Signals of Positive Selection in Humans?.


  1. Great post Larry. I think the false positive rate and algorithms is one of the biggest problems underlying this whole issue. There were a few similar talks at Evolution 2010 on this issue, and I myself work on a related issue (Functional Divergence). Luckily with Functional Divergence the question being asked is whether differences observed are evolutionarily significant, and not necessarily whether they were under positive selection.

    Reframed in that light the tests for positive selection still tell us something useful in terms of changes that may have been relevant evolutionarily, whether or not they were positively selected. The big problem is that while these algorithms make reasonable assumptions about the evolutionary process, we don't have good control datasets with enough empirical knowledge to truly test them for their accuracy. The solution, as in many cases, is to try and do very good simulations. I know of one student who spoke at Evolution 2010 who is doing just that. It is at least a step int he right direction.

  2. While "natural selection" in response to changing environments most likely played the major role in creation of the different races of humanity, it is clear that selective breeding in a relatively small number of generations can lead to profound changes in physiology. This is very evident in the domestication of farm animals and pets, and particularly in breeds of dogs.

    The relatively small differences in the genomes of humans points to an apparent origin of most human todays from a very tiny pool of ancestors about 200,000 years ago. It is likely that our species almost went extinct at that time, because there is evidence for the existence of many other related hominids even over 5 million years ago. We have made a remarkable recovery and re-shaped much of the planet as a consequence of our success as a species.

    The geographic isolation of humans and relative inbreeding has permitted certain traits to predominate in the various human races in the past. However, natural environmental factors and distance are no longer the driving factors in human evolution in the past 100 years. The human lifestyle has changed dramatically to the point where obesity and reduced fertility are common. Yet the average human lifespan has markedly increased with many elderly people in extremely good health.

    With increasing technology and an expanding population of over 6.8 billion people, there are clearly new factors at play that will drive human evolution. It is inevitable that eventually we will be genetically engineering human genomes to get rid of deleterious mutations, but very soon afterwards we will be looking to make improvements. For example, the re-introduction of gills for breathing underwater may lead to an offshoot of humanity that is a truly an aquatic ape. Survival in space and on other worlds will also produce extensive changes in human physiology. Improving the adaptability of humans to synthetic prosthetics and other devices will permit human cyborgs. These are only a few examples of what probably awaits humanity in the next hundred years.

    No! Human evolution is not slowing down. It is just getting started.

  3. Interesting what leaped out at you from the Wade article. Here's the sentence I noted, thinking "Larry will like this:"

    Most variation in the human genome is neutral, meaning that it arose not by natural selection but by processes like harmless mutations and the random shuffling of the genome between generations.

  4. So my undergrad evolution class is many, many years behind me. But assuming I remember correctly...

    ...the max speed evolution can go is the mutation rate - i.e. no selection "deleting" anything. Arguably, we humans are under less selective pressure these days, and therefore should be evolving faster now than before.

    Or do I have that wrong?