Overview

Health risks and ethical dilemmas are inherent to using gene therapy for human enhancement. Sports is currently the only field where active measures are taken against genetic enhancement, known as gene doping. Present gene doping detection methods like nested PCR and the privacy invasive whole genome sequencing lack versatility to detect the plethora of genetic differences or would take up to 37 years per Olympic games respectively. Therefore, we developed a comprehensive package consisting of a cheap and upscalable gold nanoparticle based prescreening, a novelly designed, produced and tested fusion protein for targeted sequencing and a self-learning algorithm for data analysis to anticipate future gene doping development. In this way, we not only provide Doping Laboratories, but more broadly also Bloodbanks and Food Safety Institutes around the world with a versatile, efficient and reliable method to ensure a healthy and secure future.

Figure 1. Timeline of gene doping use and development in society.

1. Gene Doping: The Threat

Gene doping is a true problem. It is the first doping that could not only find interest in the athlete population, but also in the rest of society. Thereby it is not only a global, but especially an intergenerational problem, given the possibility of (accidental) germ line infections, as well as the extension it finds in the designer baby concept. These phenomena and their acceptance in society will largely shape our future, making a strong focus on and the debate about gene doping absolutely indispensable.

Due to lacking figures on actual gene doping use, unfortunately we cannot say gene doping is happening for sure. However, in this case the question should not be whether it is already happening. Is should be whether we can afford not to worry. That is what we wanted so solve, because responsible innovation starts with anticipation.

In the meantime many (conventional) doping cases keep being revealed. Doping use is not only a risk for athlete health, but also has a big impact on his or her competitors. We interviewed Moniek Nijhuis, a former professional swimmer, who participated in many European and World Championships as well as in the Olympic Games, and who won many medals. In the video she tells about her experiences with doping and what it did to her when two years after a competition it was discovered that one of her competitors had been using doping. This is why we are here with our project ADOPE, since every athlete deserves a fair chance.

2. Current Gene Doping Detection

Up until now there have no detection methods been actively implemented. However, there have been several research groups working on the topic.

2.1 Nested PCR

Earlier this year, an Australian research group published about one of the first detection method for gene doping. This method as most methods relies on a technique called nested polymerase chain reaction or nested PCR. Nested PCR is PCR modified to reduce non-specific amplification due to unexpected primer binding. To do this, nested PCR requires two sets of primers that are used in two successive rounds of PCR. Here, the second primer set allows for amplification of a secondary target within the product produced in the first round.
However, gene doping consists of a plethora of possible changes in a plethora of possible genes. Currently more than 200 genes have been identified that are involved in the development of physical performance (Moran et al. 2017). Together, they form a technical challenge that requires enormous versatility in the detection method design. This is where nested PCR, requiring 4 known primers per target site is hard to extend and where we want to improve.

2.2 Whole Genome Sequencing

A more extensive alternative to nested PCR is whole genome sequencing as an approach to tackle gene doping. One of the laboratories working on the use of this technique against gene doping is the group of professor Hidde Haisma at Groningen University. With the current technologies however, sequencing all genomes of the participating athletes in the Olympic games would cost up to 37 years. By this time, all athletes will already be out of sports. Therefore, efficiency is an important design requirement that we improve on.

3. ADOPE: the Ultimate Solution

Our product has been shaped under the influence of many stakeholders as can be read more about on the Integrated Human Practices page. In this way we integrated all design requirements in our comprehensive gene doping method that has the potential to combat gene doping now and in the future.

Figure 1.

4. ADOPE’s Impact on Lives

4.1 Versatility

In gene doping detection it is important to be able to detect many genetic differences in a plethora of performance enhancing genes. For our fusion protein we initially considered Zinc fingers and TALEN’s to target the exon-exon junctions, as these have relatively high on-target scores. However, they are not versatile enough. Therefore, we decided to use a dxCas9 with flexible guide RNA libraries to find the gene doping sites effectively. In this way we do not only guarantee versatility within gene doping detection, but might also soon improve viral detection, fetal DNA screening and food safety maintenance. These are only a few examples of areas where we have identified interested parties as can be read more about on the entrepreneurship page.

4.2 Accuracy and Reliability

Limiting the risk for false positives has been one of our most prominent focuses, because false accusations can ruin athlete’s lives. Therefore, we adhere much value to the build-in double verification. After a prescreen, we use sequence verification to provide athletes with more direct proof. As the Oxford Nanopore MinIon device that we use has an 8% misread percentage, we simulated actual read-outs and made sure the separation between gene doping and misread normal DNA is clearly present. In this way we have an optimally accurate and reliable gene doping detection method.

4.3 Efficient, High throughput, Limiting costs

As became apparent from an initial visit to the doping authority, initial development costs are not the big problem. It is the costs that are inherent to the up-scaling that matter to them. Therefore, we made sure our method is efficient, low cost, and high throughput. Professor Hagan Bayley from Oxford University, cofounder of Oxford Nanopore Technologies, pointed us at the value of barcoding and subsequent multiplexing to enhance the sequencing efficiency. Also, we decided to add our gold nanoparticle prescreen based on the initial visit to the Dutch Doping Authority. In this way we reduce the amount of samples that will pursue, reducing costs. Lastly, our fusion protein makes targeted sequencing possible that not only significantly speeds up the analysis, but also greatly reduces the amount of data that needs to be stored.

4.4 Anticipation

Doping knows a long history of development. As a rat race, detection methods have for a long time been following up on developments in doping use. We think we should not be naïve in the sense that this time we are first, since there are certainly clues that gene doping is already happening. However, gene doping is different from more conventional doping in the sense that it is highly versatile. Therefore, we thought of an inventive method to anticipate further developments within gene doping. This we do through our self-learning algorithm for gene doping sequence classification as is described on our software tool page. We challenged engineers at the Cyber Security Week to hack our method in order to become as strong as we can be against gene doping. In this way we are the first to openly use collaborative strengths in the race against doping.
We do realize that inventive and knowledgeable people could potentially mess with our system, as with almost every detection method. Therefore, we asked all stakeholders we interviewed how they would circumvent our method in order to map our weaknesses. Most directed us into RNA injections in the samples. This might indeed complicate our targeted sequencing. However, then they would need to have the specific knowledge of the actual guide RNA’s that are used, they must have access to the blood sample afterwards and this will not circumvent our prescreen. When the prescreen is positive and the sequence verification negative, a second blood sample might be requested, which fits in with the standard practice in doping testing.

4.5 Biosafety

As a team, we believe biosafety is more than biocontainment, which we achieve by having a cell free method. Therefore, we created an infographic on how safety has been integrated throughout our project, from start to finish.

Figure 2. Infographic Safety-by-Design.

References Infographic

4.6 Sensitivity and Minimal Invasivity

We made an extensive model of the response of the human body to different administration methods of gene doping. From this model we derived the most likely ways in which athletes would be gene doping and establish a reachable detection window. Given the peak effect of the red blood cell reaction as an example, we expect athletes to be using gene doping in between competitions. This means, we should include our test in the out of competition testing, which is generally mostly urine based. Therefore, we prospect future application of ADOPE in urine testing. According to Casadio et al. 2017, concentrations of up to 138 ng/µL can be found in urine samples, providing us with confidence that in the near future our method might be extended to urine samples. [sensitivity is around….]

4.7 Privacy in Genomic Data Analysis

We believe that guaranteeing athletes’ privacy in gene doping detection is one of the most important values to take into consideration within our design. However, after talking to athletes it appeared that on average only very little value is attributed towards genomic privacy as clean sports is seen as most important. During our surveys it appeared to be a general tendency not to care about genomic privacy. Only 22.6% of the general Dutch population thinks we should be utterly careful about the genomic privacy of athletes based on a 181 respondents.

Hence, why is genomic privacy important? From one of our initial talks with Professor Dimeo it became apparent that there is not only a threat of hackings on databases. In addition, all the people that work in the regulatory cycle that get to see the results also give rise to an opportunity for blackmailing athletes that Professor Dimeo is afraid of.

The fear of genome database hacking is not fiction. Scandals are already happening. Big companies known for their ancestry analyses for example have their databases hacked, leaving 92 million people with their genetic code up for grabs (Shaban, June 5 2018). It happened this year. And if they are not hacked, some even sell this data to pharmaceutical companies that can in turn make a lot of money with it.

Why should you or any athlete be afraid of hacking? Apart from the fact that companies can earn much money with your genetic code, in time your genetic information can be used against you. An example is in health and life insurances. If this continues, in many countries only the good risks can still have a cheap insurance once their health insurer knows their genetic disease predispositions. The same applied for job and sponsoring offers and loans and mortgages. This can lead to large social problems and possibly unethical situations that we will then need to find other solutions for. For our method at least it means we highly value athlete privacy.

4.8 Current data storage and safety

Recently it became apparent that around the 2018 Olympic Winter Games in PyeongChang, the medical information of 250 athletes from 30 countries who have a dispensation to use certain medication was hacked (NOS, October 5 2018). The year before, the same hackers had already obtained the information of several other athletes. We make sure that no more data is stored than is absolutely necessary, thereby guaranteeing that no personal information will be leaked by a hack.