Decreased accuracy of forensic DNA mixture analysis for groups with lower genetic diversity

Maria Flores; Evan Ho; Cara Ly; Niquo Ceberio; Miguel Guardado; Kamillah Felix; Hannah Mariko Thorner; Matt Paunovich; Chris Godek; Carina Kalaydjian; Rori V Rohlfs

doi:10.1016/j.isci.2024.111067

Decreased accuracy of forensic DNA mixture analysis for groups with lower genetic diversity

iScience. 2024 Sep 28;27(11):111067. doi: 10.1016/j.isci.2024.111067. eCollection 2024 Nov 15.

Authors

Maria Flores^{1

2}, Evan Ho^{1

3}, Cara Ly^{1

4}, Niquo Ceberio^{1

5}, Miguel Guardado⁶, Kamillah Felix^{1

7}, Hannah Mariko Thorner⁸, Matt Paunovich¹, Chris Godek⁹, Carina Kalaydjian⁹, Rori V Rohlfs^{1

7}

Affiliations

¹ San Francisco State University, Department of Biology, San Francisco, CA 94132, USA.
² University of California, Los Angeles, Department of Molecular, Cell and Developmental Biology, Los Angeles, CA 90095, USA.
³ University of California, Santa Cruz, Department of Ecology and Evolutionary Biology, Santa Cruz, CA 95060, USA.
⁴ Yale University, Department of Genetics, New Haven, CT, USA.
⁵ Northern Arizona University, Department of Applied Physics and Materials Science, Flagstaff, AZ, USA.
⁶ University of California, San Francisco, Biological and Medical Informatics Graduate Program, San Francisco, CA 94143, USA.
⁷ University of Oregon, Department of Data Science, Eugene, OR 97403, USA.
⁸ George Washington University, Department of Forensic Sciences - Forensic Molecular Biology, Washington, DC 20007, USA.
⁹ San Francisco State University, Department of Mathematics, San Francisco, CA 94132, USA.

Abstract

Forensic investigation of DNA samples from multiple contributors has become commonplace. These complex analyses use statistical frameworks accounting for multiple levels of uncertainty in allelic contributions from different individuals, particularly for samples containing few molecules of DNA. These methods have been thoroughly tested along some axes of variation, but less attention has been paid to accuracy across human genetic variation. Here, we quantify the accuracy of DNA mixture analysis over 83 human groups. We find higher false inclusion rates for mixtures with more contributors and for groups with lower genetic diversity. Even for three-contributor mixtures where two contributors are known and the reference group is correctly specified, false inclusion rates are 1e-5 or higher for 36 out of 83 groups. This means that, depending on multiple testing, some false inclusions may be expected. These false positives could be lessened with more selective and conservative use of DNA mixture analysis.

Keywords: Genetics; Genomics.

Grants and funding

T32 GM148332/GM/NIGMS NIH HHS/United States