Artificial Intelligence in Genetic Testing

 

The human species has transformed, from an average life expectancy of about 30 years to living in a vast technological world, creating machines rivaling even human intelligence. French sociologist Emile Durkheim argued that as society develops, people seek independence and, accordingly, specialize in different fields, explaining modern-day specialization in ever-expanding fields [1]. Artificial intelligence is humanity’s newest breakthrough. It has swiftly crept into many of today’s scientific fields. Artificial intelligence refers to a machine or system with the ability to simulate human intelligence. It applies human-like abilities, such as learning and perceiving, for problem-solving and pattern recognition. AI functions by being trained on considerable amounts of data. Following that, testing takes place on the capabilities of AI [2]. For example, GPT-2, the predecessor of Chat-GPT released in 2019, was trained on 1.5 billion parameters; however, its successor GPT-3 was trained on 175 billion parameters [3].

With AI’s capabilities of accelerated and more exact genome sequencing, it could prove to be a useful tool in the field of genetics and medicine. AI could be used to analyze the genetic makeup of large groups of people, leading to the discovery of common genetic markers as well as rare mutations, while also helping us discover groups at risk of certain diseases. With AI’s ability to analyze large amounts of data, it could easily analyze an individual’s genes as well as take their environment and lifestyle into account. In cancer research and diagnosis, AI-powered machine models can be used to detect biological markers linked to cancer. This gives patients a chance to receive an earlier diagnosis. Discoveries such as these can be remarkably useful in the treatment of these patients who are at risk but may not know it. [4]. In 2021, researchers experimented using GEM eCDSS on groups of patients diagnosed with rare genetic diseases from the Rady Children’s Hospital, the Boston Children’s Hospital, the Hudson Alpha Institute for Biotechnology, and other universities and institutes. GEM showed astonishing results, recognizing over 90% of the markers for these rare genetic diseases. [5] Conversely, another recent example of artificial intelligence being used in disease detection is the AI-MARVEL tool, also known as AIM. Researchers at the Jan and Dan Duncan Neurological Research Institute developed this model specifically to test for genetic diseases. Reportedly, it had a precision of 98%, identified 57% of diagnosed cases, and outperformed all benchmarked methods. AIM achieved this by receiving high-quality samples curated by the ABMGG. [6] Many other examples similar to AIM or GEM exist, and they exhibit the promising future AI could give us.

Despite the fact that AI is a highly capable tool, many issues arise when it comes to the use of AI in genetic testing. Respect for autonomy and non-maleficence are some of the fundamental ethical principles in modern medicine. Consequently, these principles must also apply to the use of AI for genetic testing. Another important issue is transparency. The AI machine models used for genetic testing must be transparent. If not made transparent, this could lead to difficulty identifying errors and using algorithmic bias for diagnosis, which has the capability of completely putting them at risk [7]. Moreover, informed consent is a process of communication between patients and healthcare providers, where patients have the right to be made aware of the full extent of their diagnosis, treatment, etc. Informed consent as a process must also be taken into consideration if AI should be used for genetic testing. In the EU, GDPR was enforced as a way to protect the privacy of EU citizens, and in the USA, an organization known as GINA exists directly working against misuse and discrimination against individual’s genetic makeup. Organizations and laws like this are important when it comes to the usage of AI [8].

In conclusion, AI is a modern invention of mankind, showcasing incredible pattern recognition and learning. As established, it has shown to have exceptional potential to positively impact humanity’s future. Moreover, it can alter the course of our society, shifting us toward a brighter path where patients receive quicker and more adequate help. However, taking into consideration that, since its inception, discussions on ethical and privacy concerns have always been present, it is evident that many issues must be addressed. With enough dedication and diligence, AI can be used to revolutionize genetics, ultimately helping those struggling with genetic diseases.

Citation list

1. Émile Durkheim. 1997. The Division of Labor in Society. Translated by W. D. Halls. New York: The Free Press. (Originally published in 1893.)
2. Partha Pratim Ray. ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope, Internet of Things and Cyber-Physical Systems, Volume 3, (2023). doi : 10.1016/j.ejca.2023.02.030
3. Yongjun Xu, Xin Liu, Xin Cao, Changping Huang, Enke Liu, Sen Qian, Xingchen Liu, Yanjun Wu, Fengliang Dong, Cheng-Wei Qiu, Junjun Qiu, Keqin Hua, Wentao Su, Jian Wu, Huiyu Xu, Yong Han, Chenguang Fu, Zhigang Yin, Miao Liu, Ronald Roepman, Sabine Dietmann, Marko Virta, Fredrick Kengara, Ze Zhang, Lifu Zhang, Taolan Zhao, Ji Dai, Jialiang Yang, Liang Lan, Ming Luo, Zhaofeng Liu, Tao An, Bin Zhang, Xiao He, Shan Cong, Xiaohong Liu, Wei Zhang, James P. Lewis, James M. Tiedje, Qi Wang, Zhulin An, Fei Wang, Libo Zhang, Tao Huang, Chuan Lu, Zhipeng Cai, Fang Wang, Jiabao Zhang. Artificial intelligence: A powerful paradigm for scientific research. The Innovation. Volume 2, Issue 4. (2021). doi : 100179
4. Vilhekar RS, Rawekar A. Artificial Intelligence in Genetics. Cureus. 2024 Jan 10;16(1):e52035. doi: 10.7759/cureus.52035. PMID: 38344556; PMCID: PMC10856672.
5. Francisco M. De La Vega, Shimul Chowdhury,  Barry Moore,  Erwin Frise, Jeanette McCarthy, Edgar Javier Hernandez,   Terence Wong, Kiely James, Lucia Guidugli, Pankaj B. Agrawal, Casie A. Genetti, Catherine A. Brownstein, Alan H. Beggs, Britt-Sabina Löscher, Andre Franke, Braden Boone,  Shawn E. Levy, Katrin Õunap,   Sander Pajusalu, Matt Huentelman, Keri Ramsey, Marcus Naymik, Vinodh Narayanan, Narayanan Veeraraghavan, Paul Billings, Martin G. Reese, Mark Yandell &  Stephen F. Kingsmore. Artificial intelligence enables comprehensive genome interpretation and nomination of candidate diagnoses for rare genetic diseases. Volume 13, article number 153, (2021) doi : https://doi.org/10.1186/s13073-021-00965-0
6. Texas Children's Hospital. "AI-MARRVEL: New AI Tool to Diagnose Genetic Disorders." Texas Children's Hospital. Accessed February 25, 2025.
7. Coghlan S, Gyngell C, Vears DF. Ethics of artificial intelligence in prenatal and pediatric genomic medicine. J Community Genet. 2024 Feb;15(1):13-24. doi: 10.1007/s12687-023-00678-4. Epub 2023 Oct 5. PMID: 37796364; PMCID: PMC10857992.
8. Farhud DD, Zokaei S. Ethical Issues of Artificial Intelligence in Medicine and Healthcare. Iran J Public Health. 2021 Nov;50(11):i-v. doi: 10.18502/ijph.v50i11.7600. PMID: 35223619; PMCID: PMC8826344.
9. https://www.news-medical.net/health/AI-Powered-Genomic-Analysis-Revolutionizing-the-Detection-of-Genetic-Mutations.aspx
10. https://scopeblog.stanford.edu/2022/06/10/using-ai-to-find-disease-causing-genes/