Genomics and athletic performance: an emerging discipline that is not yet ready for society

Abstract

Genomics of athletic performance is an emerging discipline with a high degree of controversy. With the existing level of evidence, it is both premature and highly risky to exploit current human genomics knowledge to predict exercise and sports performance or enhance existing training methodologies. Until more solid evidence on the influence of genomic variants in athletic performance becomes available, accompanied by regulatory approved genome-guided recommendations, all genetic associations should be restricted from general public access as commercial services, since genomic markers cannot per se predict athletic performance for talent identification, resistance to injuries or the ability to recover from them. Evidently, the complex interplay of genetics with other physical, physiological and even psychological and mental characteristics to produce a world-class athlete is still not understood.

Main text

It is well known that athletic performance is dependent upon genetic factors by approximately 50%. In particular, endurance and power-related characteristics are within a range of 44–68% and 48–56%, respectively [2, 4]. The remaining determinants of athletic performance are attributed to non-genetic factors, included but not limited to different training methods, biological gender, diet, ethnicity, and psychological factors, which in addition to environmental factors that either modulate gene expression or epigenetic modifications impact on the performance among athletes. From the above, it is well understood that performance in endurance and power sports is a highly complex issue, in which genetics plays a significant but not the only role. Specifically, only a small fraction of the more than 250 variants in more than 140 genes are believed to significantly impact on athletic performance [3].

In our recent meta-analysis of 11501 endurance and power athletes and 42881 control subjects (54382 subjects in total; Psatha et al., [6], which was based both on a thorough literature search and a stringent statistical analysis, described in detail in the Methods section, we have shown that there is no statistically significant association between genomic variants and athletic performance either for endurance or power sports.

We have read with interest the letter by Flück and coworkers (2025), who arraigned us of being biased and intentionally attempting to “…undermine the application of genetics in athlete development”, claiming that we used a poorly developed approach, resulting in misleading conclusions. Their criticism lies on the fact that confounding factors such as ethnicity and biological gender were not taken into consideration in our analysis and according to them, this was the weakest point in our analysis.

The role of biological gender and ethnicity is indeed established in sport performance. However, in some studies the biological gender is not at all considered, other studies included only male or female athletes while others analyzed both male and female athletes as a single group, making it extremely challenging to meta-analyze studies considering biological gender as a confounding factor! Although we have recognized this limitation in our study in numerous places in the text and discussed it in detail, which was appreciated by the referees and also admitted by Flück and coworkers (2025) in their letter, we opted to focus solely on ACE and ACTN3 variants association with athletic performance, as (a) some of the studies that were included in our meta-analysis only alluded on the athletes’ biological gender and the same was true for the athletes’ ethnicity, without implicating and/or considering in the actual analysis either biological gender or ethnicity, and (b) if we were indeed able to include biological gender and ethnicity in our analysis, then the number of studies would have been very limited to allow a proper meta-analysis to be performed.

In our view, which coincides with the views of several other genomics experts, the weakest points in all these genome-wide association studies are not the confounding factors but the study design itself. Firstly, these limitations mostly stem from imprecise phenotypic measurements as classification of sports, since there are no uniform criteria to assign an athlete to the power or endurance category. Furthermore, determining an athlete’s performance as elite, sub-elite, etc., cannot be objectively determined, and credible criteria for such classification are currently also not available, not to mention the strong influence of epigenetics as well as environmental or other factors (e.g. weather conditions, stumbling on a hurdle, etc.) in determining the overall outcome of a world-class competition, let alone determining what a world-class competition really is (e.g. Olympics, World or European championships, Diamond League, etc.). In our study, we chose not to reclassify the athletes and to follow both the athlete and sport classification reported in each of the study included in the meta-analysis, as this would introduce bias in our analysis. This answers to the unfortunate criticism of Flück and coworkers of our analysis being biased!

Coming back to the ethnicity cofounding factor, there are several sports teams that seriously consider genetic tests results when making direct coaching recommendations [8]. However, possible associations between genes and athletic performance at the study population level revealed by GWAS are not consistent at an individual’s level, therefore genetic testing alone cannot conclusively confirm or rule out an individual’s athletic performance [1]. This has a direct ethical implication, since many genetic testing laboratories sell genetic testing services of specific genomic variants with the promise to elucidate the athletic performance potential of individuals, especially the youth, as there is significant risk of misleading interested individuals from false positive or false negative results [5, 6].

It appears that Flück and coworkers (2025) have selectively read our article and seems that were they who interpreted our conclusions in a biased and subjective fashion. Moreover, we believe that their last paragraph that falsely accused us of drawing intended conclusions was prejudiced, as we simply reported our findings from an straightforward meta-analysis of the existing literature on this topic, thoroughly described and diligently performed, to answer a clearly set research question by interpreting our results in an objective fashion, leading to our conclusions that sports genetics is an emerging discipline, which while promising, is still in its infancy and not ready yet for society.

With the existing level of evidence, it is both premature and highly risky to exploit current human genomics knowledge to predict exercise and sports performance or enhance existing training methodologies, which is also outlined by Varillas-Delgado and coworkers [7] in their recent comprehensive review. Until more solid evidence on the influence of genomic variants in athletic performance becomes available, accompanied by regulatory approved genome-guided recommendations, all these reported genetic associations should remain in the investigational sphere, and restricted from general public access as commercial services, since genomic markers cannot per se predict athletic performance for talent identification, resistance to injuries or the ability to recover from them, and the complex interplay of genetics with other physical, physiological and even psychological and mental characteristics to produce a world-class athlete is still not understood.