Epigenetic Clocks: A New AI Model for Estimating Biological Age from DNA Methylation
Epigenetic mechanisms, specifically DNA methylation, play a key role in aging. However, the specific biological processes are not entirely understood. Despite the variability in research perspectives, the functional decline associated with aging remains a significant focal point of scientific interest.
DNA methylation-based biomarkers have shown promise in predicting age-related changes across different DNA sources. One specific model, known as XAI-AGE (Explainable AI for AGE), integrates previously identified hierarchical biological information into a neural network model for predicting biological age based on DNA methylation data. This model aligns with the hierarchy of biological pathways, similar to Elmarakeby’s tool.
XAI-AGE outperforms earlier prediction models and matches deep learning models in accurately predicting biological age from DNA methylation. It is especially effective in whole blood and blood PBMC tissue types. This model offers easy interpretation of results across tissues, age groups, and cell line differentiation.
In conclusion, this new neural network model for age estimation, XAI-AGE, provides precise and interpretable predictions based on DNA methylation data. It offers insights into the mechanisms connected to aging and provides a foundation for generating hypotheses. The model has proven to be a significant advancement in the research of AI and aging.
For more information, you can check out the research study here. And don’t forget to join their Telegram Channel!