A New Robust Epigenetic Model for Forensic Age Prediction.

Area of Science:

• Forensic Science
• Genetics
• Biotechnology

Background:

• Forensic DNA phenotyping aims to predict individual characteristics from biological samples.
• Accurate age prediction is crucial as aging affects many externally visible traits.
• Existing epigenetic clocks are highly accurate but not feasible for typical forensic labs.

Purpose of the Study:

• To develop a simplified epigenetic clock for age prediction suitable for forensic applications.
• To identify reliable DNA methylation markers for age estimation in the Italian population.
• To validate a machine learning model for forensic age prediction.

Main Methods:

• Analysis of 42 methylation markers from five genes in 330 Italian subjects.
• Development of a prediction model using a machine learning approach (ridge linear regression).
• Validation of the best-performing model on an independent sample of 83 individuals.

Main Results:

• A ridge linear regression model with eight methylation markers was identified as the best predictor.
• The model achieved a median prediction error of 4.5 years on an independent validation set.
• The study successfully validated an epigenetic model for age prediction in the Italian population.

Conclusions:

• A simplified epigenetic clock using DNA methylation markers has been developed and validated for age prediction.
• The model shows promise for forensic casework, with a median error of 4.5 years.
• Limitations include potential reduced accuracy for advanced ages in forensic contexts.