Sergey Brin, one of the founders of Google, once shared a big secret to the public. Through a genomic test he took, he found out he actually carries the LRRK2 gene mutation. The LRRK2 gene is a “blueprint” that can increase his risk of getting Parkinson’s disease in old age. However, instead of giving up, he intended to do something. He dared to invest a large amount of funds to combine genetics with Artificial Intelligence. The goal is clear, he wants to dissect millions of existing cellular data to predict exactly when the disease gene will “turn on” and how to muffle it before the first symptoms appear. The question is: Can technology really “see” our health future even though it is still 20 years away? The key lies in the machine’s ability to dissect millions of codes making up the human body. Welcome to a new era, an era where your biological health can be read with mathematical precision.
The Limitations of the Human Brain in Reading an Ocean of Data
Let us look at the reality. Your body is like an information library that is very large and complete in size. Did you know, that the human DNA chain consists of more than 3 billion base pairs (genetic letters). If some information inside one of your body cells wants to be printed in a book, the pile could beat the height of a skyscraper! In each of those genetic informations, there are millions of epigenetic “switches” (DNA methylation points) that respond to everything, especially our lifestyle. As explained in the previous article, this “switch” can turn on or turn off disease genes, depending on our daily lifestyle. Finding the pattern of damage among millions of genetic information is like looking for a needle in a haystack. Our brains, no matter how smart we are, certainly have biological limitations. We ourselves will not be able to process, remember, and calculate trillions of data points manually to predict exactly when an organ will start to fail to function. Therefore, that is where AI and machine learning work.
“In the era of longevity, the biggest challenge of the medical world is not the lack of patient data, but the inability of the human brain to find the pattern of aging in that ocean of data.”
Bioinformatics and the Birth of Machine Learning
This is the turning point where Artificial Intelligence (AI) can play a huge role in Longevity Medicine. To be able to perform very complex computations, there is a new branch of science called Bioinformatics. A combination of molecular biology with computer science to help computation. Inside Bioinformatics, scientists use one of the derivatives of AI called Machine Learning. This is far different from a regular calculator, because Machine Learning is designed to learn by itself from new data. Imagine AI as a super smart librarian capable of reading millions of your DNA books in a matter of seconds. AI works by:
- Recognizing Hidden Patterns: AI scans millions of your DNA methylation points and compares them with the history data of other patients.
- Detecting Wear and Tear: AI can find out what kind of DNA methylation patterns, that can cause micro inflammation, which ones trigger brain shrinkage, and which ones can increase the risk of heart blood vessel blockage. So it is very accurate.
- Predicting: Instead of just saying you are “healthy today”, AI can be designed to make predictions of your organ aging starting 10 to 20 years into the future.
Mathematical Accuracy That Changes Medicine
Is the prediction from AI accurate? Based on the latest scientific research, the answer is Yes, AI predictions are very convincing. AI integration is no longer science fiction, but also becomes a new standard for diagnostics, especially in Longevity Medicine. This is affirmed in a study in the journal Ageing Research Reviews (Zhavoronkov et al., 2020), which revealed that current AI technology has been able to identify human aging biomarkers in real-time with an accuracy that is impossible to achieve by conventional medical statistics or clinical predictions. In fact, this AI technology has created a very specific measuring tool. A publication in the journal Aging (Galkin et al., 2021) introduced DeepMAge, an Epigenetic Clock purely built using Deep Learning algorithms. This study revealed that AI or Artificial Intelligence can suppress the margin of error in predicting someone’s biological age with a very high level of precision, far exceeding the analytical ability of human doctors.
“AI does not guess your future. It merely calculates the medical probability of what you are doing to your body today.”
Are You Ready to See Your Future Trajectory?
Many people feel afraid to take AI-based genetic or epigenetic tests. There is a kind of psychological anxiety: “What if the results are bad? It is better I do not know at all.” However, let us change our mindset. Reading your aging prediction with the latest technology is not something like an absolute “curse” that cannot be changed. The question now is: If you have the opportunity to see the blueprint and aging prediction of your body organs precisely, so you can do the right planning to prevent them from getting more damaged, would you dare to take that opportunity?
REFERENCES:
- Novakovsky, G., Dexter, N., Libbrecht, M. W., Wasserman, W. W., & Mostafavi, S. (2023). Obtaining genetics insights from deep learning via explainable artificial intelligence. Nature Reviews Genetics, 24(2), 125-137.
- Galkin, F., et al. (2021). DeepMAge: A methylation aging clock developed with deep learning. Aging (Albany NY), 13(7), 9852-9869.
- Zhavoronkov, A., et al. (2020). Artificial intelligence for aging and longevity research: recent advances and perspectives. Ageing Research Reviews, 64, 101176.
