In Tibet, human hypoxia response was a major evolutionary adaptation to high elevations (probably acquired via introgression from archaic hominins called Denisovans). In the Andean highlands, in contrast, where the potato was domesticated, there are adaptations for starch digestion but not for hypoxia.
The Andean highland people also demonstrate the widely recognized reality that highlands and other inaccessible places tend to be refugia for populations wiped out elsewhere, while lowlands are more vulnerable to population replacement.
The genetic split date between low and high elevation populations in the Andean mountains, however, ca. 7150 BCE to 6150 BCE, is surprisingly recent for a territory that modern humans had very likely colonized by 12000 BCE.
The peopling of the Andean highlands above 2500m in elevation was a complex process that included cultural, biological and genetic adaptations. Here we present a time series of ancient whole genomes from the Andes of Peru, dating back to 7,000 calendar years before present (BP), and compare them to 64 new genome-wide genetic variation datasets from both high and lowland populations.
We infer three significant features: a split between low and high elevation populations that occurred between 9200-8200 BP; a population collapse after European contact that is significantly more severe in South American lowlanders than in highland populations; and evidence for positive selection at genetic loci related to starch digestion and plausibly pathogen resistance after European contact.
Importantly, we do not find selective sweep signals related to known components of the human hypoxia response, which may suggest more complex modes of genetic adaptation to high altitude.John Lindo, et al., "The genetic prehistory of the Andean highlands 7,000 Years BP though European contact" bioRxiv (July 31, 2018).