This web page was produced as an assignment for an undergraduate course at Davidson College.

"The Athletic Gene "

Popular Press v. Scientific Literature:

Is ACTN3 responsible for determining an individual's athletic ability?

Elite sprint athlete Marian Jones. Image from BBC Sport Online.

Popular Press Article: Gene linked to athletic performance. Click here to view article.

According to this Yahoo article, Australian researchers have identified a gene that is directly related to a person’s athletic ability. Does this mean we can determine how good of an athlete someone will be based on a genetic test? Does this make you want to find out if you “have the gene?” Maybe researchers can develop a procedure to “correct” a person who does not have this gene?

The article covers a study conducted by Yang et al. in 2003 in which they studied the allele frequencies of elite athletes in order to determine if the gene ACTN3 is linked to athletic performance. A functional copy of the gene ACTN3 gives rise to alpha-actinin-3, a protein found in fast muscle fibers. The article explained that there are two known alleles of ACTN3: an R allele and an X allele. The R allele encodes a functional copy of the protein while the X allele does not produce the protein. The DNA of more than 300 top athletes was screened to determine which version of the gene each person had. The study revealed that 95 percent of the elite sprint athletes had at least one copy of the R allele while 50 percent of the sprint athletes had two copies of the R allele. 76 percent of the endurance athletes were found to have one copy of the R allele and only 31 percent had two copies of the R allele. The results for the endurance athletes were similar to the results for the control group (percentages not specified in article). Only 5 percent of the sprint athletes had a copy of the X allele while 24 percent of endurance athletes were found to have two copies of the X allele. Interestingly, 18 percent of the control group was found to have the X allele. The statistics stated in the article are displayed in Table 1.

ACTN3 Allele
% Sprint Athletes
% Endurance Athletes
% Control Group
R allele (1 copy)
similar to endurance athletes
R allele (2 copies)
similar to endurance athletes
X allele (1 copy unless otherwise specified)
24 (2 copies of X allele)

Table 1. Percentages of ACTN3 R allele and ACTN3 X allele for elite, endurance and control groups. Note that the percentage of X alleles under endurance athletes is for two copies of the X allele, not one. The popular press article was slightly confusing in that it combined statistics for one and two copies of the same allele.

The statistics show that the R allele appears with higher frequency in sprint athletes. Is this just coincidence, does the R allele make someone a better athlete, or does the R allele provide someone with a sprint/power advantage? While the article began by inferring the gene for athletics was discovered, it never directly draws any conclusions about the R allele being present in “better athletes.” The article indicates that ACTN3 has an affect on the type of muscle fibers (rather than whether a person is a good or bad athlete) and quotes researcher Kathryn North as saying “I hypothesize that absence of alpha-actinin-3 means that an individual’s muscles are more ‘slow’ in character, and better suited for endurance activities.” The article also negates its bold claim at the beginning about genetic screening with a quote from the chairman of UK sport saying that it “would only ever give an indication, albeit a potentially valuable one, as to a child’s athletic promise.” The ACE gene is also mentioned as having an influence on athletic ability which suggests that more than one gene plays a role in determining athletic ability. The popular press article started out as a sensational story about the gene for athleticism, but ended with some confusion and many unanswered questions.

Background on ACTN3 Gene

The alpha-actinins are a family of proteins that bind to actin. ACTN2 and ACTN3 are genes found in humans that encode skeletal muscle proteins that bind to actin, but they are differentially expressed, both spatially and temporally (Yang et al, 2003). ACTN2 is expressed in all skeletal muscle fibers, whereas ACTN3 is expressed only in fast-twitch myofibers. Alpha-actinin-3 (the protein encoded by ACTN3) plays a role in an individual’s ability to generate speed and quick, powerful movements. Recently, North et al. found that approximately 16 percent of the world population is homozygous for a nonsense mutation (R577X) in ACTN3, resulting in a deficiency of alpha-actinin-3 (OMIM, 2003). Alpha-actinin-2 is most likely able to compensate for the lack of alpha-actinin-3 (North et al., 1999), but the ACTN3 gene has been highly conserved through evolution which suggests that it has an function independent from the function of ACTN2. Surprisingly, the deficiency of alpha-actinin-3 varies across populations: 25 percent of Asian populations are deficient, 18 percent of Europeans, and only less than 1 percent of the African Bantu populations are deficient. The different alleles may have an influence on the differences in muscle function and individuals with alpha-actinin-3 may have an advantage for sprint/power activities (Yang et al., 2003).

Scientific Press: ACTN3 Genotype is Associated with Human Elite Athletic Performance by Yang et al. 2003. Click here to view paper.

Yang et al., 2003 studied the allele frequencies of the R577X locus in groups of sprint and endurance athletes to establish a correlation between the ACTN3 genotype and a sprint/power athletic advantage. They found that elite sprint athletes have higher frequencies of the 577R allele than the controls which suggests that "the presence of alpha-actinin-3 has a beneficial effect on the function of skeletal muscle in generating forceful contractions at high velocity."

This study genotyped 436 white control individuals (all unrelated and from different sources) and 429 elite white athletes from 14 different sports that had competed for Australia at an international level. A subset of the elite athletes (107 individuals) was classified as specialists for “sprint/power” sports including track, short distance swimming and cycling, and speed skating. A subset of 194 different individuals were classified as endurance specialists and participated in sports such as cross country skiing, rowing, and long distance running, swimming and cycling. The paper used the log-linear modeling approach (as described by Huttley and Wilson, 2000) to determine individual alleles and genotype frequencies between groups. They found that the allele frequencies for the control group did not significantly differ from the allele frequencies found in previous control groups, so this indicates that their control group is representative of the overall white population. The allele frequencies did not deviate from Hardy-Weinberg equilibrium and there were no significant differences in genotype frequencies between males and females.

The athlete group as a whole had similar genotype frequencies to the control population; significant difference was not noted until the athlete group was divided into sprint/power athletes and endurance athletes (for both males and females). The sprint athletes had a lower frequency for the XX genotype (6 % for sprint athletes, 18 % for control group), a higher frequency for the RR genotype (50% for sprint, 30% for control) and a lower frequency for the heterozygous RX genotype (45% for sprint, 52% for control). The endurance athletes had a higher frequency of the XX genotype (24% for endurance, 18% for controls). The athlete group as a whole had similar allele frequencies to the control group because the frequencies of the sprint athletes and the control athletes compensated for one another. The data in the paper is displayed in Figure 1 (the figure is taken from Yang et al., 2003). A slight deviation from Hardy-Weinberg equilibrium was seen between female sprint athletes and female endurance athletes while no difference was seen in males. No female sprint athletes were found to be XX. The authors hypothesized that the ACTN3 genotype has a different effect on performance for males and females. When males exercise, the androgen hormone response improves performance, so the effect of ACTN3 on sprint and power ability may not be as significant in males as it is in females. The data suggests “the ACTN3 577R allele provides an advantage for power and sprint activities.” Because most elite sprint athletes possessed at least one copy of the 577R allele, the authors predicted that even one copy of the 577R allele improves performance (gain-of-function).

Figure 1. Sprint athletes have higher genotype frequencies for ACTN3 RR than do control or endurance athlete groups. The bars on the graph are color coded according to the legend. The sprint athlete groups all show higher genotype frequencies for RR than any of the other groups, while the endurance athlete groups all have higher genotype frequencies for XX than any of the other groups. It is interesting to note that very few of the sprint athletes were found to have the XX genotype. Figure was taken from Yang et al., 2003.

Since the 577R allele provides an obvious phenotypic advantage, one would wonder why the 577X allele has been so highly conserved throughout evolution. One possibility the authors suggested is that the advantages of the 577R allele only become apparent under the most extreme conditions (i.e. elite athletic competition) so that the 577R allele only provides an advantage during few circumstances. In this case, the 577X allele would then provide a neutral phenotype and have become established in the population during random genetic drift. A second hypothesis is that the 577X allele has become established in the population due to an advantageous polymorphism in close proximity to the ACTN3 gene, although this speculation is unlikely to due genetic analysis. A third possibility is that the 577X allele has been selected for due to phenotypic advantages of its own. The data showed that the endurance athletes had a higher than normal frequency of the XX genotype, suggesting that the 577X allele may increase endurance performance. If the 577X and 577R allele each provide a unique phenotypic advantage under different circumstances, then it seems likely that they have both been highly conserved in the population due to balancing selection. The authors are currently conducting research to determine the origin of the X allele and the selective pressure placed on R577X.

The authors speculate that alpha-actinin-3 may provide an advantage for sprint and power activities by promoting fast-twitch muscle fiber formation or changing the glucose metabolism during athletic training. Alpha-actinin-3 may facilitate the absorption or transmission of force during muscle contraction, thereby minimizing damage. Alpha-actinin-3 has been shown to bind to the gluconeogenic enzyme fructose-1,6-bisphosphatase, to the glycogen phosphorylase amorphin, and to the calsarcins. The authors have developed a knockout mouse for Actn3 to better investigate how alpha-actinin-3 affects muscle function.

The authors made a note that the ACE gene has also been linked to athletic advantages; different ACE allele frequencies have been observed for both sprint and endurance athletes. ACE encodes the angiotensin-converting enzyme and plays a role in blood pressure regulation (OMIM, 2003). Many other genetic loci have been identified that influence fitness and athletic performance, but ACTN3 is the first gene identified that encodes a protein for the skeletal muscle. The authors propose that there is a "trade-off" between sprint/power abilities and endurance abilities and that each individual is predisposed towards one of these traits. It seems likely that balancing natural selection between the two alleles of ACTN3 has helped to maintain this genetic variation.


When one first reads the popular press article, one would think that the athletic gene has been found and now all people can be genetically screened to determine if they are a “good” athlete or a “bad” athlete. The article starts off with a sensational first paragraph and immediately presents the question as to whether children can be screened at birth to determine their athletic ability. The article uses a controversial idea to engage readers and spark interest, even if the idea may not be based on scientific fact. The scientific article tells the reader the purpose of the study, describes the research and displays the data. The scientific paper does not attempt to sensationalize any of the data to keep people reading, but rather focuses more on interpretation of the data.

Overall, the popular press article presented the data accurately and in such a way that the general public could understand it. The experiment was summarized fairly well and the basic concept of two alleles for the ACTN3 gene was described but the article was also a little misleading and confusing at times. I thought it was helpful for the article to include the overall percentages of each group that possessed a certain allele; the data were easy to understand when presented in this manner. Although the article starts out with a radical idea, it concludes with remarks showing that the case may not be as cut and dry as originally stated. The article does a good job of conveying the main idea, but people who are familiar with scientific publications will find the paper to be much more informative. The scientific paper explains the research in a much more detailed (and biological) manner.

I thought a very interesting aspect of the scientific paper was the speculations as to why both the 577R and 577X allele have been conserved over time. It was also interesting to learn about the differences in allele frequencies between males and females. The popular press article did not bring up the Hardy-Weinberg equilibrium and was also a little misleading in that is dubbed the 577R allele as “good” and the 577X allele as “bad.” The article focused mainly on the 577R allele and did not explain that the 577X allele has been evolutionarily conserved and has been associated with an increase in endurance performance. The only way a reader would find out this piece of information would be to read the scientific paper.


After reading the popular press article, I was excited that the ACTN3 gene might be able to partially explain an elite athlete’s amazing talent. However, as soon as I thought about it, I realized how unrealistic it was to think that athletic ability could be based solely on one gene. Many many factors help determine what kind of an athlete somebody is. Physical factors such as height and weight and mental factors such as determination and work ethic all are part of what makes an athlete and each factor is determined by a multitude of genes (and also influenced by the environment). Also, trying to decide who is a “better” athlete is sort of a moot point; different body types are better suited for different sports, so how could you determine who is a better athlete between a gymnast and a hockey player or between a downhill skier and a tennis star? It’s like comparing apples and oranges.

In conclusion, a popular press article may outline an exciting breakthrough in science, and some of it may even be based on real data, but to learn the real story it is much more helpful to read it straight from the source, the scientific publication. The popular press is a good way to make the public more aware of findings in science, but sometimes the information will be dramatized or over-simplified. Don’t believe everything you read; make sure to analyze the data from the experiment. As much as everyone would like all traits to follow Mendelian genetics, it is rare to find one gene that will control one trait only and nothing more. Everything in biology is part of a complicated and interconnected web, so nothing is as simple as it may first appear.

Related Sites

Click here to view the amino acid sequence of ACTN3 for Homo sapiens.

Click here to view the ACTN3 entry in OMIM's database.

Click here to learn more about the gene ACE from OMIM's database.

Works Cited

[OMIM] Online Mendelian Inheritance in Man. 2003. ACTN3. <>. Accessed 2003 Sept. 7.

Yahoo! Headlines. 2003 Sept. 2. Gene linked to athletic performance. <> . Accessed 2003 Sept. 7.

Yang et al. 2003. ACTN3 Genotype is associated with human elite athletic performance. Am. J. Hum. Genet. 73: 627-631. <>. Accessed 2003 Sept. 7.



Back to Sarah's Homepage for Genomics, Proteomics and Bioinformatics.


Davidson College

Davidson College Biology Department

Davidson College Genomics Homepage

Sarah's Molecular Biology Homepage

Please send questions and comments to Sarah Baxter at Thanks!