TY - JOUR
T1 - A new implementation of a semi-continuous method for DNA mixture interpretation
AU - Alfieri, Jacob
AU - Coble, Michael D.
AU - Conroy, Carole
AU - Dahl, Angela
AU - Hares, Douglas R.
AU - Weir, Bruce S.
AU - Wolock, Charles
AU - Zhao, Edward
AU - Kingston, Hanley
AU - Zolandz, Timothy W.
N1 - Funding Information:
The authors would like to thank Professor Klass Slooten (VU University Amsterdam; Netherlands Forensic Institute) for sharing the MixKin software and helpful discussions in the development of the PopStats SC mixture program. This is publication number 22.09 of the FBI Laboratory Division. Names of commercial manufacturers are provided for identification purposes only, and does not imply endorsement of the manufacturer, or its products or services by the FBI. The views expressed are those of the author(s) and do not necessarily reflect the official policy or position of the FBI or the U.S. Government. This work was supported in part by award 2020-DQ-BX-0022 from the US National Institute of Justice to the University of Washington. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of their organizations.
Funding Information:
The authors would like to thank Professor Klass Slooten (VU University Amsterdam; Netherlands Forensic Institute) for sharing the MixKin software and helpful discussions in the development of the PopStats SC mixture program. This is publication number 22.09 of the FBI Laboratory Division. Names of commercial manufacturers are provided for identification purposes only, and does not imply endorsement of the manufacturer, or its products or services by the FBI. The views expressed are those of the author(s) and do not necessarily reflect the official policy or position of the FBI or the U.S. Government. This work was supported in part by award 2020-DQ-BX-0022 from the US National Institute of Justice to the University of Washington . Points of view in this document are those of the authors and do not necessarily represent the official position or policies of their organizations.
Publisher Copyright:
© 2022 The Authors
PY - 2022/12
Y1 - 2022/12
N2 - A new calculation module within the PopStats module of the CODIS software package, based on the underlying mathematics presented in the MixKin software package, has been developed for assigning the Likelihood Ratio (LR) of DNA mixture profiles. This module uses a semi-continuous model that allows for population structure and allelic drop-out and drop-in but does not require allelic peak heights or other laboratory-specific parameters. This new implementation (named SC Mixture), like MixKin, does not specify or estimate a probability of drop-out. Instead, each contributor to a mixture has an independent drop-out rate, and the probability of the mixture profile for a specified proposition concerning the contributors is integrated over the range of possible drop-out rates. The allelic drop-in rate and the population structure parameter, theta, used by the software are specified by the user. The user can examine up to five contributors to a mixture, however, conditioning on assumed contributors and limiting the number of unknowns in both numerator and denominator hypotheses greatly improves performance. We report results from an extensive validation study performed for ten mixtures with each of one (single source), two, three, four, or five contributors, with four combinations of drop-in rate and a population structure parameter. Each mixture was run as a complete profile or with the random removal of alleles to simulate drop-out. All 1620 combinations were evaluated with PopStats, MixKin, and LRmix and considerable consistency was found among the results with all three packages.
AB - A new calculation module within the PopStats module of the CODIS software package, based on the underlying mathematics presented in the MixKin software package, has been developed for assigning the Likelihood Ratio (LR) of DNA mixture profiles. This module uses a semi-continuous model that allows for population structure and allelic drop-out and drop-in but does not require allelic peak heights or other laboratory-specific parameters. This new implementation (named SC Mixture), like MixKin, does not specify or estimate a probability of drop-out. Instead, each contributor to a mixture has an independent drop-out rate, and the probability of the mixture profile for a specified proposition concerning the contributors is integrated over the range of possible drop-out rates. The allelic drop-in rate and the population structure parameter, theta, used by the software are specified by the user. The user can examine up to five contributors to a mixture, however, conditioning on assumed contributors and limiting the number of unknowns in both numerator and denominator hypotheses greatly improves performance. We report results from an extensive validation study performed for ten mixtures with each of one (single source), two, three, four, or five contributors, with four combinations of drop-in rate and a population structure parameter. Each mixture was run as a complete profile or with the random removal of alleles to simulate drop-out. All 1620 combinations were evaluated with PopStats, MixKin, and LRmix and considerable consistency was found among the results with all three packages.
KW - DNA mixtures
KW - Forensic DNA
KW - Mixture interpretation
KW - Probabilistic Genotyping
KW - Semi-continuous method, Likelihood Ratio
UR - http://www.scopus.com/inward/record.url?scp=85132798637&partnerID=8YFLogxK
U2 - 10.1016/j.fsir.2022.100281
DO - 10.1016/j.fsir.2022.100281
M3 - Article
AN - SCOPUS:85132798637
SN - 2665-9107
VL - 6
JO - Forensic Science International: Reports
JF - Forensic Science International: Reports
M1 - 100281
ER -