Gene in Focus: SOD2August 20, 2016
How Your DNA Impacts Your DietSeptember 1, 2016
You might have heard a few weeks ago that we released our first peer reviewed paper looking at part of the DNAFit test – the Peak Performance Algorithm. In case you missed this, you can find the some of the media coverage here and you can download the research paper itself here. This research makes DNAFit the first sporting genetic testing company to provide research showing the utility of their product.
The idea behind the research was to validate the algorithm DNAFit use to determine the best type of training for each person. We all know intuitively that we respond different to the same training; if you’ve ever had a training partner I’m sure its fair to say that you both did not see exactly the same improvements. The difference in improvements between individuals is partly genetic, and research in recent years has focused on identifying genes that can play a role in training response. One of the most well researched of these is ACTN3, which creates a protein found in fast twitch muscle fibres. Research has shown that a person’s ACTN3 genotype can have an impact on how much fast-twitch muscle fibre they have, as well as having an association with testosterone levels. It has also been shown to affect the size of improvement from a strength training programme. ACTN3 is not just the only gene to have an impact on how well you respond to a training programme; many more genes have been identified, with differing strengths of supporting evidence. DNAFit have strict criteria determining which genes we test for, and all genes in our panel must have a minimum of three peer reviewed papers supporting their inclusion. As such, we test for 15 genes that have been reliably shown to influence training programme response. We put the results from these 15 genes into our algorithm, and this algorithm tells us each individuals power and endurance percentage, which in turn can guide training programme design.
The big question, though, is whether this process works, and that’s what we attempted to show in this paper. We recruited 67 male subjects in two separate experiments. In experiment 1, we used 28 male athletes aged 18-20, who were members of British Universities & Colleges Sport (BUCS) teams in various sports. In study 2, we used 39 male soccer players aged 16-19. The two study groups were from different centres with different coaches involved. Each subject had at least 6 months and a maximum of 2.5 years of resistance training experience, so they were not beginners. This is important as beginners tend to see really big improvements no matter what training they do, which can sometimes cause results to be misleading. The athletes were randomly assigned to genetically matched or mismatched training. Meanwhile all subjects underwent a DNAFit test, and were scored as either power athletes (>50% power score) or endurance athletes (>50% endurance score). Matched athletes would be power athletes doing high-intensity training or endurance athletes doing low-intensity training; mismatched athletes would be power athletes doing low-intensity training or endurance athletes doing high-intensity training. Neither the athletes or coach knew whether they were matched or mismatched – this is called double blinding, and is important to ensure that athletes don’t try harder if they think they are in the right group, or try less if they think they are in the wrong group.
All the subjects followed a resistance training programme for 8 weeks, with a minimum of one, but ideally two, training sessions per week, alongside their typical sports training. They all did the same 6 exercises, which were deadlifts, pull-downs, front squats, dumbbell flat press, step-ups and vertical jumps. The only difference between the groups were the number of sets and repetitions they did. Those in the high intensity group did 10 sets of 2 repetitions on all exercises. Those in the low-intensity group did two weeks of 3 sets of 10 repetitions, then three weeks of 3 sets of 15 repetitions, then 3 sets of 20 repetitions for the final three weeks.
Before and after this 8-week training programme, all subjects underwent some tests. These tests were a measure of power called the countermovement jump (CMJ) and a measure of endurance called the Aero3. Both of these tests were selected as they are tests of performance; being able to jump higher in useful in many sports, as is having better endurance.
So, what did we find? The headline news is that those subjects doing genetically matched training saw greater improvements than those doing mismatched training. The average improvement in CMJ across both studies for those in the mis-matched group was 2.6% (standard deviation
±5.3), whilst for those in the matched group it was 7.4% (4.9). For the Aero3 test, the average improvement for mismatched athletes was 2.3% ( ±3.1), whilst for matched athletes it was 6.2% (3.2). All of these results were statistically significant. The graph below (figure 1) illustrates this:
Figure 1 – The mean percentage improvements in both CMJ and Aero3 for both matched and mismatched athletes.
The researchers then looked at how well each individual improved with exercise, and split subjects into three groups; high, moderate and low responders. For the CMJ, 82.6% of high responders were following genetically matched training, whilst only 18.2% of the low responders were. For the Aero3 test, 86.4% of high responders were following matched training, compared to 18.2% in the low responder group. This means that after 8 weeks of resistance training, the odds of achieving a more favourable outcome in the CMJ were 21 times greater (p<0.0001) for those doing matched training compared to the mismatched group. For the Aero3, the odds were even better, with subjects doing matched training 28 times more likely (p<0.0001) to have a favourable outcome than those doing mismatched training. Figure 2 (below) shows the training response type according to matched or mismatched training.
Figure 2 – High, moderate and low responders for both CMJ and Aero3 according to matched or mismatched training.
Because of space constraints within the journal, we weren’t able to report all the individual results for our subjects – fortunately we can do this here for the first time. Figures 3.1 and 3.2 below show the changes in CMJ score for those athletes doing matched and mismatched training.
Figure 3.1 – CMJ Improvement (%) compared to DNAFit Power Score in mismatched athletes [mean improvement = 2.6% (standard deviation
Figure 3.2 – CMJ improvement (%) compared to DNAFit Power Score in matched athletes [mean improvement = 7.4% (
Visually, you should be able to see that those doing genetically matched training were generally seeing much greater improvements in their CMJ performance. You can see each subjects’ individual CMJ score on the above graphs, along with their power percentage. Remember, a power percentage score of less that 50% would make that person an endurance athlete, and about 50% would make that person a power athlete according to our criteria. There is a very slight trend towards those with a greater power percentage having a slightly larger improvement in the CMJ, but overall this effect is small.
The same is true for improvements in Aero3 (below – figures 4.1 and 4.2). Again, you can see that matched athletes saw much greater improvements in Aero3 compared to the mismatched group:
Figure 4.1 – Improvements in Aero3 compared to DNAFit Power Score in mismatched athletes [mean improvement 2.3% (
Figure 4.2 – Improvements in Aero3 compared to DNAFit Power Score in matched athletes [mean improvement = 6.2 (
Finally, there has been a great deal of recent discussion focusing on the use of genetic testing for talent identification. DNAFit have always maintained that genetic tests cannot be used as a talent ID tool, and a recent consensus statement in the British Journal of Sports Medicine supported this. The results from our study illustrate this notion; there was no significant difference between the two groups in the performance tests at baseline, as shown below in figures 5.1 and 5.2:
Figure 5.1 – Initial CMJ Score stratified for Genotype Group. The endurance group had a mean CMJ of 35.24cm (SD
±6.06) and the power group had a mean of 35.27cm ( ±5.11).
Figure 5.2 – Initial Aero3 Score stratified for Genotype Group. The endurance group had a mean score of 2106.10m (
±220.02), and the power group had a mean of 2134.54m ( ±216.35).
The actual baseline values for CMJ are shown below in figures 6.1 and 6.2 – the range of values was similar in both the matched and mismatched group, this is a good sign of the effectiveness of the randomization, there was no bias in either group that could explain the results of the tests
Figure 6.1: The individual baseline CMJ values in the mismatched group, i.e. the initial CMJ test values before training began
Figure 6.2: The individual baseline CJM values in the matched group.
Finally, we saw a range of power genotype scores. The lowest power score we had was 10.3% (indicating an endurance potential of 89.7%). The highest power score 80%, and the mean score was 47.4%. The frequency of each genotype score is detailed below in figure 7:
Figure 7 – Frequency distribution of power genotype scores.
So what does this all mean for you? If you’re training, you can now use a genetic test to increase your chances of improvement from that training programme. If you’re a power responder, as predicted by the DNAFit test, then your best chance of success comes with high-intensity resistance training; high load and a low number of repetitions. If you’re an endurance responder, then your best training programme is low-intensity resistance training; moderate load with a high number of repetitions. This is the case whatever are your goals – you may be an endurance-responder sprinter, or a power-responder marathon runner! If you’re a personal trainer or a coach, this knowledge can enable you to maximize your clients or your athletes’ improvements from training, which can be a real game changer for you. The next steps for DNAFit are to replicate these results in other studies, in order to continue to show the use of a DNAFit test when it comes to designing a training programme. We are also continuing to look for new genes and SNPs we can test for, which we will use in order to improve the service DNAFit offer