Use of Artificial Intelligence with Epigenetic Biomarkers
/
Zhavoronkov, Aging Volume 11, Issue 22. November 25, 2019
Aging is the principal risk factor in many chronic diseases such as cancer, cardiovascular, and metabolic & neurological diseases. There is therefore great interest in the development of accurate age biomarkers that can be targeted and measured to track the effectiveness of therapeutic interventions and other lifestyle adjustments.
Right now epigenetic biomarkers can fairly accurately determine a subject’s biological age and even predict the chronological age of the individual, but they can’t provide an individual with actionable results on which activities are beneficial for their health and which are not.
Enter, Deep Neural Networks & AI Based Aging Clocks
In recent years though, artificial intelligence (AI) and machine learning (ML) have started pushing into sectors that were previously considered out of their reach. These computer systems have now made their way into aging research, where they’ve spawned the creation of next-generation ‘deep aging & longevity clocks’.
In this recent paper, Alex Zhavoronkov et al. show how these deep learning algorithms can help advance aging research by:
establishing causual relationships in non-linear systems
being utilized for the identification of novel therapeutic targets
evaluating the efficacy of various interventions
predicting health trajectories & mortality
These techniques can be utilized to establish causality by using age predictors, which can then be used to identify the most important targets in a specific stage of a disease, and even personalized interventions. This approach can hopefully one day be used to identify pathological changes that transpire during aging, and identifying what personalized therapeutic interventions are required to return an individual to the state resembling their optimal biological age.