The Future of AI in Genetic Testing: An Analysis of AI’s Potential Role & Impact

AI is undoubtedly a pillar of future advancements in healthcare. Despite concerns over bias and security risks, AI has already been integrated into healthcare, and genetic testing is only one area AI is particularly able to impact. In 2022, 129,624 genetic tests were developed and submitted to the genetic testing registry [1]. That is a testament to the crucial role genetic testing has in identifying diseases, assessing risks, and providing early treatment plans. AI serves as a catalyst for genetic testing’s potential by improving efficiency, refining test results, and increasing accessibility.

Healthcare providers aim to make genetic testing more widespread [2]. This, however, is difficult when interpreting test results remains exclusive to specialized professionals who are few and far between [3]. A study surveying clinical geneticists showed that appointment wait times for genetic testing have significantly increased, with 39% of new non-emergency patients waiting over three months [3]. The study also revealed that most clinical geneticists are over 50, and a quarter plan to retire soon, illustrating this field’s staffing shortages [3]. 

To address this issue, AI is capable of simplifying and streamlining the process, relieving the burden on clinical geneticists and allowing them to meet the needs of patients. In fact, AI-powered systems are already being used to detect mutations in genes. A recent study on copy-number variants – variations in DNA segment copies between individuals – and exome sequencing – a technique examining protein-coding regions for disease-related mutations – discovered that machine learning generates far more accurate results in detection [4]. This suggests AI has the potential to improve efficiency, reducing the need to undergo multiple tests to confirm results. 

Standard genetic tests typically focus on a single gene mutation or a specific variant to diagnose a suspected condition [5]. Contrarily, whole exome sequencing looks across one’s entire DNA [5]. While this sounds useful, whole exome sequencing is relatively expensive and tends to miss certain diagnoses [6]. In a study on exome sequencing in detecting rare and undiagnosed diseases, 33% of 54 patients were not successfully addressed using exome sequencing alone [6]. Luckily, AI has access to huge databases on diseases and genetic mutations, meaning it can extract and compare data at faster rates and produce more comprehensive analyses. This makes assessing multiple genes or complex traits easier.

Genetic testing is only impactful if patients understand their results and receive timely treatment. If the physician is not equipped to make this possible, the value of genetic testing is lost. Genetic counseling is a critical aspect of patient care that AI can enhance. A research article found that physicians often have difficulty interpreting and communicating test reports, especially for rare genetic diseases [7]. AI can process complex reports and present them in a simpler way, lifting the pressure on physicians who are less knowledgeable in this field. With America’s low health literacy rates and direct-to-consumer genetic tests rising in popularity, AI can further foster accessibility by increasing the readability of test results for patients as well [8, 9]. AI can then complement reports with images and videos, catering to different learning styles. A study analyzing ChatGPT’s ability to improve plastic surgery postoperative instructions proves this by concluding that readability for instructions increased without compromising safety or accuracy [10]. 

It is important to consider the limitations of AI as well. Since AI is still developing, security and privacy concerns are a prime obstacle. Genetic testing companies like 23andMe have already experienced data breaches where nearly 7 million customers were exposed [11]. AI having access to sensitive patient information could result in more vulnerability. Bias and mishaps with AI also pose an issue. The expectation that AI can provide more accurate analyses is dependent on how well AI can evolve in the near future. Additionally, while AI can simplify genetic counseling, it cannot replace it. The human touch of healthcare is key to uphold, especially when patients receive life-changing diagnoses and need support to cope with the news.

AI is transforming modern technology in ways we have never imagined. Implementing AI into genetic testing reduces the time and resources required in this procedure, ultimately increasing the affordability of these tests. When the burden on doctors to counsel patients and on patients to understand their results is mitigated, we are one step closer to achieving equitable and accessible healthcare. Pushing the limits of genetic testing and allowing AI to aid with better detection is great, but ensuring everyone can reach these benefits to begin with is even more incredible.

Works Cited

1. Halbisen, A. L., & Lu, C. Y. (2023). Trends in Availability of Genetic Tests in the United States, 2012-2022. Journal of personalized medicine, 13(4), 638. https://doi.org/10.3390/jpm13040638 

2. De La Vega, F.M., Chowdhury, S., Moore, B. et al. (2021). Artificial intelligence enables comprehensive genome interpretation and nomination of candidate diagnoses for rare genetic diseases. Genome Med 13, 153. https://doi.org/10.1186/s13073-021-00965-0 

3. Jenkins, B. D., Fischer, C. G., Polito, C. A., Maiese, D. R., Keehn, A. S., Lyon, M., Edick, M. J., Taylor, M. R. G., Andersson, H. C., Bodurtha, J. N., Blitzer, M. G., Muenke, M., & Watson, M. S. (2021). The 2019 US medical genetics workforce: a focus on clinical genetics. Genetics in medicine : official journal of the American College of Medical Genetics, 23(8), 1458–1464. https://doi.org/10.1038/s41436-021-01162-5 

4. Danecek, P., Gardner, E. J., Fitzgerald, T. W., Gallone, G., Kaplanis, J., Eberhardt, R. Y., Wright, C. F., Firth, H. V., & Hurles, M. E. (2024). Detection and characterization of copy-number variants from exome sequencing in the DDD study. Genetics in medicine open, 2, 101818. https://doi.org/10.1016/j.gimo.2024.101818 

5. National Library of Medicine. (n.d.). What are the different types of genetic tests? MedlinePlus. https://medlineplus.gov/genetics/understanding/testing/types/ 

6. Burdick, K. J., Cogan, J. D., Rives, L. C., Robertson, A. K., Koziura, M. E., Brokamp, E., Duncan, L., Hannig, V., Pfotenhauer, J., Vanzo, R., Paul, M. S., Bican, A., Morgan, T., Duis, J., Newman, J. H., Hamid, R., Phillips, J. A., 3rd, & Undiagnosed Diseases Network (2020). Limitations of exome sequencing in detecting rare and undiagnosed diseases. American journal of medical genetics. Part A, 182(6), 1400–1406. https://doi.org/10.1002/ajmg.a.61558 

7. Scheuner, M. T., Edelen, M. O., Hilborne, L. H., Lubin, I. M., & RAND Molecular Genetic Test Report Advisory Board (2013). Effective communication of molecular genetic test results to primary care providers. Genetics in medicine : official journal of the American College of Medical Genetics, 15(6), 444–449. https://doi.org/10.1038/gim.2012.151 

8. Kumari, A. (2024, November 29). Direct-to-Consumer Genetic Testing Markets. BCC Research. https://blog.bccresearch.com/direct-to-consumer-genetic-testing-markets 

9. Hickey, K. T., Masterson Creber, R. M., Reading, M., Sciacca, R. R., Riga, T. C., Frulla, A. P., & Casida, J. M. (2018). Low health literacy: Implications for managing cardiac patients in practice. The Nurse practitioner, 43(8), 49–55. https://doi.org/10.1097/01.NPR.0000541468.54290.49

10. Zhang, A., Li, C. X. R., Piper, M., Rose, J., Chen, K., & Lin, A. Y. (2024). ChatGPT for improving postoperative instructions in multiple fields of plastic surgery. Journal of plastic, reconstructive & aesthetic surgery : JPRAS, 99, 201–208. https://doi.org/10.1016/j.bjps.2024.08.065 

11. Helmore, E. (2023, December 5). Genetic testing firm 23andMe admits hackers accessed DNA data of 7m users. The Guardian. https://www.theguardian.com/technology/2023/dec/05/23andme-hack-data-breach