Rare variants and human genetic diversity

You may have read that there is more genetic variation within major ethnic groups than between ethnic groups. That this precludes the possibility of group differences is Lewontin's Fallacy (explained in pictures).

Deep sequencing of the human genome, which reveals rare variants (here, defined as those found in fewer than 0.5 percent of the population), shows that there is actually more variation between groups than within groups. (So what you may have been taught in school is not true -- sorry, that's how science works sometimes.) The figure below, from this July 6 Science article, shows that over 50 percent of rare genetic variants are found in African populations (which have greater genetic diversity) but not in European populations. About 41 percent of all rare variants are found only in Europeans and not in Africans, and only 9 percent of the variants are common to both groups.

These rare variants are likely recent mutations. Unsurprisingly, they differ in populations that have been geographically separated for tens of thousands of years.

The excerpt below is from this paper, in the same issue of Science.

Because rare variants are typically the result of recent mutations, they are expected to be geographically clustered or even private to specific populations. Using a measure of variant sharing between two samples (7), we found that for common variants, any two European populations appear to be panmictic, whereas for rare variants, European populations show lower levels of sharing (fig. S7). In general, the level of sharing depends on geographic distance, with the dependence increasing substantially with decreasing allele frequency (fig. S8). The Finnish population shows substantially lower levels of sharing with other European populations than predicted by geographic distance, which is consistent with hypotheses of a historical Finnish demographic bottleneck (23). Levels of rare variant sharing are even lower when comparing populations from distinct continents. Thus, catalogs of rare variants will need to be generated locally across the globe. 

We found substantial variation in the total abundance of variants across populations, even within Europe (Fig. 3 and fig. S7D), which is likely due to demographic history. In particular, we observed a north-south gradient in the abundance of rare variants across Europe, with increased numbers of rare variants in Southern Europe and a very small number of variants among Finns, who had about one third as many variants as southern Europeans.

There seems to be a correlation between latitude and prevalence of rare variants. Could mutation rates vary due to average temperature?