on here you get a lot of apparently good American runners and they seem to always advise that we not 'race' our tempos, that we hold back. Canova argues that this is the difference, that Kenyans run 12-40 minute tempos very very fast, and that they do a lot of progression runs where they finish their miles very fast. Malmo, who did his morning recovery runs at an average of well under 6 minutes per mile (in other words, a tough guy whose easy runs were all well within a minute of his marathon pace), would still run 4 mile tempos at over 5 minute mile pace, if I remember correctly. Keep pushing the tempo runs to insane paces, while modulating your pace far more by doing extremely slow recovery runs (so that the tempos don't destroy you), and maybe you reach far greater heights. And then you add on the long, early years of aerobic development at altitude and you have a big engine for the running.
As for the genetics question, no evidence has ever been shown to support that, but look at it this way. All human beings are descendants of the Rift Valley (e.g. Ethiopia and Kenya). So, any genes that developed in the Rift Valley peoples by, say, 70,000 to 140,000 years ago would have been passed on, with alterations due to random mutation, adaptation, and sexual selection, to those who lift the Rift Valley and populated the rest of Africa and the other continents. So, why would Rift Valley residents have genes for distance running that the rest of the world didn't have? It seems unlikely that the distance running genes suddenly popped up after the migrations, since distance running was, according to a Harvard study, a primary driver in very early human evolution.