Gluteus

The abstract of the paper that explains the test:

https://www.ncbi.nlm.nih.gov/pubmed/20133430

In a nutshell:
"VO2max responses to endurance training can be predicted by measuring a approximately 30-gene RNA expression signature in muscle prior to training."

hhnngg1

A dirty little (actually BIG) secret in molecular biology:

Those Affymetrix gene chips for DNA and RNA are quite variable in their data precision and reliability. When they were invented, they were hailed as the be-all end-all of genomics, but in practice, it's closer to a random crapshoot than anything else. I'm actually not exaggerating this at all - if you try and submit a paper where they main claims are based mostly on Affymetrix chip data, no matter how convincing, you will get instantly rejected by all the top journals (Cell, Science, Nature) unless you have much more substantive proof backing up the finding. This is largely why that paper, despite it's pretty ambitious claim that they can explain a whopping 30% of VO2 with simple gene analysis, is not published in one of these top journals. If Affymetrix data was rock-solid, it would have shot right to the top, as that sort of data if rock solid would completely change the way we train and study top athletes right now today.

I've seen so many PhD proposals by colleagues based on Affymetrix data that seemed so exciting and promising because it seemed to correlate "X" gene with this great phenomena, be totally shot down when you actually do the validation work. In our lab and others we collaborated with, it was known that you could almost always get Affymetrix data to be a supporting figure of any phenomena you chose to study - just repeat the chip analysis a few times until the variability in the results gives you the result you want. I know this sounds incredible cynical, but it's absolutely true. There are no major papers that rely solely on Affymetrix data because of this well known variability. You MUST back up your findings with prospective experimental trials to get reviewers to believe you.

I actually fully believe that this type of genetic analysis will absolutely become integral to elite athlete training and selection in the future, but we're nowhere near that level today. Papers like this one are interesting, but until they prove that their Affymetrix data is more than random variation, we've got a long way to go before that day.

pgjackson

There aren't a lot of scientific studies on sports performance. But if you coach sports it doesn't take long to realize that some people just have "it". They are naturally better. If they work at it they will become better at it than their peers. They have a gift...just like some people are amazing artist, musicians or mathmaticians. They can just do things others can not. There is something in their genetic make-up that makes them unusually well suited for the activity. Why can't we all sing like Ray Charles? I'm pretty sure Ray didn't have a lot of professional training, but he is amazing. No amount of training is going to get me a sub 4-minute mile. I will never bench press 500lbs. Bradley Wiggins, no matter how hard he tries, will never win a marathon. It's not his thing. Lebron James, as great an athlete he is, will never win a professional bike race. And 99.9% of us will never be elite in any athletic activity because of our genetic makeup. We could train real hard and be the best local guy...but to be elite it takes more than just hard work and determination.