Fresh bone samples gave the most reliable age estimates

What the study found: The study found that DNA methylation analysis can generally be used to estimate age-at-death from fresh bone material, but reliability was lower under forensically relevant conditions. Among the five bone types tested, only petrous bone showed age estimation errors that were statistically significantly different from at least one other bone type.
Why the authors say this matters: The authors indicate that age estimation tools for bone samples are being developed for forensic use, including for unknown corpses or skeletons. The study suggests that bone type and challenging environmental exposure may affect how well methylation-based age estimation works.
What the researchers tested: The researchers used the VISAGE enhanced age estimation tool for bones on freshly collected bone samples from 30 individuals. They analyzed five bone types from each individual: rib, femur, clavicle, iliac crest, and petrous bone. They also exposed femur samples to burying, submersion in water, or burning, and analyzed two actual casework samples.
What worked and what didn't: The tool produced high overall age estimation errors in the fresh bone samples, with a mean absolute error (MAE) of 9.79 years and a mean error of -7.25 years. A linear transformation between predicted and chronological age improved the overall MAE to 7.03 years. In the mock casework samples, only 12 samples produced a result, with an MAE of approximately 10-15 years after transformation; for the two actual casework samples, age estimates were possible but differed from chronological age by 10 and 29 years.
What to keep in mind: The available summary does not describe additional limitations beyond the observed errors and the difficulty of obtaining reliable results from environmentally affected bone. The findings are based on 30 individuals, five bone types, mock casework exposures, and two actual casework samples.

Key points

• Fresh bone samples generally allowed age-at-death estimation using DNA methylation analysis.
• Only petrous bone showed statistically significantly different age estimation errors compared with at least one other bone type.
• The VISAGE bone tool had a high overall MAE of 9.79 years in the fresh samples.
• A linear transformation improved the overall MAE to 7.03 years.
• Mock casework exposures produced limited results, and the two actual casework samples differed from chronological age by 10 and 29 years.