Making decisions in missing person identification cases with low statistical power.

Area of Science:

• Forensic Genetics
• Population Genetics
• Bioinformatics

Background:

• DNA database searches are crucial for missing person identification.
• Complex cases with few or distant relatives present challenges due to low statistical power.
• The identification of abducted children in Argentina serves as a key example.

Purpose of the Study:

• To propose a general strategy for missing person identification using DNA database searches.
• To address challenges posed by low statistical power in complex pedigrees.
• To develop a statistical method and user-friendly software for calculating likelihood ratio thresholds.

Main Methods:

• Development of a statistical method for selecting likelihood ratio (LR) thresholds based on error rates.
• Creation of open-source, user-friendly software for computing LR thresholds and error rates.
• Application of the strategy to large-scale identification cases, exemplified by the "Missing Grandchildren of Argentina".

Main Results:

• A robust statistical method for determining appropriate LR thresholds in low-power identification scenarios.
• An accessible software tool for calculating these thresholds and associated error rates.
• A validated strategy applicable to diverse large-scale DNA-based identification challenges.

Conclusions:

• The proposed strategy enhances the accuracy of missing person identification in challenging cases.
• The developed software provides a practical tool for forensic geneticists and researchers.
• This approach offers a scalable solution for identifying individuals in mass disaster or historical cases.