Exercise is good for everyone, but it doesn't affect us all in the same ways: some people might see boosts in levels of endurance, while others benefit from better blood sugar levels. Now scientists think they may have discovered one of the reasons why.

In a new study, 654 adults with a mostly sedentary lifestyle were put through a 20-week endurance exercise program, while the levels of around 5,000 different proteins in their blood were analyzed from samples.

The researchers identified hundreds of proteins corresponding to someone's 'trainability', or how well they're going to respond to exercise – findings that could lead to more personalized and effective training regimes, as well as help experts to tackle disease.

"While groups as a whole benefit from exercise, the variability in responses between any two individuals undergoing the very same exercise regimen is actually quite striking," says Robert Gerszten, chief of cardiovascular medicine at the Beth Israel Deaconess Medical Center (BIDMC).

"To date, no aspects of an individual's baseline clinical profile allow us to predict beforehand who is most likely to derive a significant cardiorespiratory fitness benefit from exercise training."

A total of 147 proteins were linked to cardiorespiratory fitness, technically known as VO2max: it measures how much oxygen your whole body can use and transfer to your muscles. The greater the value, the fitter you are (if you've got a fancy smartwatch, it might be able to measure VO2max for you).

The team was also able to identify 102 proteins related to the change in VO2max once the exercise program had been completed. These proteins were then worked up into a score that predicted an individual's trainability – how much of a shift in VO2max the training would bring about.

Using the score, the researchers were able to pick out the study volunteers whose VO2max or cardiorespiratory fitness level wouldn't get much of a boost from the endurance exercises – useful information to have when it comes to trying to improve your fitness.

"Baseline levels of several proteins predicted who would respond to the exercise training protocol far better than any of our established patient factors," says Gerszten.

Gerszten and his colleagues also ran a separate community-based study that matched some of the identified proteins with a higher risk of early death, backing up the idea that these proteins are closely related to heart health and mortality rates.

While it's unlikely that we're going to get an exercise pill anytime soon, these are vital insights into how the human body responds to physical exercise – and how we might be able to come up with more personalized, more effective exercise routines in the future. It's the first study to look at the link between proteins and VO2max in this much detail.

The next stage is to test out the hypothesis on a larger sample of people, with more data points in terms of both factors influencing VO2max and the number of proteins measured.

While it's clear that we all respond to exercise differently, it's less clear why that happens – and this could be the start of properly answering the question.

"We now have a detailed list of new blood compounds that further inform our understanding of the biology of fitness and exercise adaptation, and predict individual responses to a given exercise regimen," says Gerszten.

The research has been published in Nature Metabolism.