Bayesian networks for DNA-based kinship analysis: Functionality and validation of the GENis missing person identification module. (December 2022)

• Record Type: Journal Article
• Authors: Chernomoretz, Ariel; Marsico, Franco; Iserte, Javier; Herrera Piñero, Mariana; Escobar, Maria Soledad; Balparda, Manuel; Sibilla, Gustavo
• Is Part Of: Forensic science international. Volume 8(2022)
• Page Start: 131
• Page End: 132
• Publication Date: 2022-12
• Subjects: Forensic genetics -- Periodicals; Forensic Genetics -- Periodicals; Electronic journals; 614.1
• DOI: 10.1016/j.fsigss.2022.10.008
• Languages: English
• ISSNs: 1875-1768

Abstract

Abstract: GENis is a recently published open-source multi-tier information system developed to run forensic DNA databases. It relies on a Bayesian Networks framework and it is particularly well suited to efficiently perform large-size queries against databases of missing individuals. In this contribution we present a validation of the missing person identification capabilities of GENis. To that end we introduce fbnet, a free-software package written in the R statistical language that implements the complete GENis functionality to perform kinship analysis based on DNA profiles. With the aid of fbnet, we could validate likelihood ratios against estimations draw with Familias and forrel (two well-recognized R packages for kinship quantification) for complex pedigrees provided by the Argentinian reference databank (Banco Nacional de Datos Geneticos, BNDG). We found that our methodological approach presented an excellent performance in terms of accuracy and computation times.