It appears, based on the interesting submitted article, that it would be more informative to report about 100 meter dash winners where they trained, and who their coaches were. Much the same could be said about marathon winners, who tend to train in a different part of the world.

After edit: the kind first reply to my comment here mentions a possible study design for looking at what might influence athletic success among athletes from one place or another. A "genetically sensitive" design for such a study would have to consider quite a few other issues, not least actual individual genes, which have been surprisingly neglected in genetic research over the years. (Rather than actually looking at genes of each individual in a study, which continues to be a generally expensive and occasionally error-prone enterprise, most previous genetic studies of human behavioral characteristics have been based on proxy assemblages of genes, by studying closely related family members, or in the best cases monozygotic twins.)

Readers who haven't seen this pair of links before, which I share fairly often here on HN because they are by world famous human behavioral geneticists, may be surprised by what they learn from reading about what twin studies can and cannot say about genetic influences on human behavior.

http://people.virginia.edu/~ent3c/papers2/Articles%20for%20O...

http://people.virginia.edu/~ent3c/papers2/Articles%20for%20O...

Meanwhile other readers have commented, and tying in what I've written (and linked to) above to what several comments here say, a common fallacy comes from assuming that "race" (as socially defined in one place or another) has anything much at all to do with what genes one would find in a particular individual. Serious geneticists are actually some of the leading scientists in suggesting that in medicine, law, and education we should pay less attention to "race" than formerly, to avoid inherent cognitive biases in explaining individual differences.