Structured incorporation of prior information in relationship identification problems

Area of Science:

• Genetics
• Forensic Science
• Evolutionary Biology
• Conservation Biology
• Genealogy

Background:

• Relationship identification problems are prevalent across diverse scientific and forensic fields.
• Existing methods for integrating prior information are often ad hoc.
• Formal modeling of prior information is needed for consistent and robust analysis.

Purpose of the Study:

• To demonstrate a formal approach for modeling and integrating various information sources for relationship identification.
• To provide a consistent framework for utilizing prior information in pedigree analysis.
• To highlight the application of Bayesian inference and available software for these problems.

Main Methods:

• Assigning prior probability distributions to pedigree sample spaces.
• Calculating DNA data likelihoods using established software.
• Applying Bayes' Theorem to compute posterior probabilities.
• Distinguishing between local and global prior information.

Main Results:

• A consistent framework for modeling prior information in relationship identification is proposed.
• The method allows for the integration of diverse data types.
• Robustness to different prior choices can be studied, with flat priors corresponding to a likelihood approach.

Conclusions:

• The proposed method offers a consistent alternative to ad hoc approaches for incorporating prior information.
• Relationship identification problems in various fields can be effectively addressed using this formal Bayesian framework.
• Freely available software can be utilized for practical implementation.