Chromosomal inversions in human evolution

A new paper by Lars Feuk et al. in PLoS Genetics is reporting on widespread gene inversions in humans. A press release at ScienceDaily announced the paper, and summarizes the results concisely: </p>

According to [Stephen W.] Scherer, prior to this research, only nine inversions between humans and chimps had been identified. Using a computational approach, Scherer's group identified 1,576 presumed inversions between the two species, 33 of which span regions larger than 100,000 base pairs--a sizeable chunk of DNA. The average human gene is smaller, only about 60,000 bases in length.
Scherer's team experimentally confirmed 23 out of 27 inversions tested so far. Moreover, by comparing the chimp genome with its ancestor, the gorilla genome, they determined that more than half of the validated inversions flipped sometime during human evolution.

And of course if several hundred inversions occurred during human evolution, you can bet that 15 percent or so of them will vary among humans:

Perhaps even more interesting than the abundance of inversions that Scherer's group unveiled was their discovery that a subset of the inversions are polymorphic--taking different forms--within humans, meaning that the human genome is still evolving. When the 23 experimentally confirmed inversions were tested against a panel of human samples, the scientists found three inversions with two alleles or pairs of genes displaying the human inversion in some people, whereas others had one allele of the human inverted sequence and one allele of the normal sequence in chimps.

Three out of 23 is 13 percent, and the paper has this to say:

It would be expected that a certain fraction of the differences found between the human and chimpanzee assemblies are polymorphic in one of the two species, but perhaps not to the extent (13%) observed in this study (Feuk et al. 2005:e56).

Not surprising at all. Considering that most human loci have genealogical coalescents within the last million years, purely neutral alleles within any given genealogical linage have around a 15 percent chance (one million out of the six-million-year divergence) of occurring after the coalescent, and therefore being polymorphic. In reality, the odds of polymorphism are even higher, depending on the number and distribution of individuals used to ascertain variants.

But are these inversions neutral? According to the paper, the polymorphic inversions occur at 5, 30, and 48 percent frequencies (no indication whether the minor allele is ancestral or derived compared to primates). Few variants that are very much rarer than 5 percent will be ascertained by HapMap. At the same time, alleles over 5 percent frequency are pretty unlikely to be deleterious. Their high frequencies might suggest that these inversion variants have themselves been positively selected.


Feuk L, MacDonald JR, Tang T, Carson AR, Li M, Rao G, Khaja R, Scherer SW. 2005. Discovery of Human Inversion Polymorphisms by Comparative Analysis of Human and Chimpanzee DNA Sequence Assemblies. PLoS Genet 1:e56. <a href=">Full text (free)</a>