The Effect of the Analytical Threshold on the Loss and Gain of Data from Single Source and Mixed DNA Samples

Document Type


Publication Date


MeSH Terms

Forensic Sciences, Alleles, DNA, DNA Methylation

Subject: LCSH

Forensic Sciences, DNA


Forensic Science and Technology


The forensic science community has been heavily scrutinized in recent years due to the release of the PCAST report. One of the methods under scrutiny is the analysis of DNA mixtures. These methods have been reassessed due to the lack of understanding, validation and methodical testing of the effects of changing the analytical threshold (AT) and the increasing use of probabilistic genotyping software programs. Cybergenetics states that their program, TrueAllele, is the most accurate statistically for DNA mixtures due to the lack of an analytical threshold in their DNA mixture assessment. By using no analytical threshold, the maximum number of peaks will be detected in a sample without data loss. Other software analysis programs allow for forensic laboratory policies that use differing analytical thresholds (e.g. 50 RFU, 75 RFU, 100 RFU) to screen for maximum confidence in allele calls compared to instrument noise. This study attempts to test the effects that the analytical threshold has on the percent loss of peaks for both single source DNA samples and DNA mixtures. Using the program GeneMarker where the analytical threshold can be manipulated, 103 mixed source and 53 single source DNA samples taken anonymously were evaluated with various analytical thresholds and the numbers of DNA fragments/peaks were tallied. For each data set, the only adjusted parameter was the analytical threshold. On average, single source samples had a 0.14% loss per threshold shift suggesting that there is little to no effect of the analytical threshold on the final number of allele calls. Conversely, mixed source DNA samples had almost twenty times that value, with a loss of data averaging 2.82%. The analytical threshold affects mixed source samples significantly more than single source samples and depending on the amount of total peaks in the DNA mixture and data quality (peak heights), there could be a substantial loss of data when analytical thresholds are used. Touch DNA samples are common types of forensic mixtures where there may be a major component but the minor contributors are not distinguishable. This could be due to data loss below the analytical threshold, reducing the number of informative loci for forensic reporting. The effect of this is to reduce clarity for number of contributors to a mixed sample and a loss in discrimination power or ability to identify the minor components in the mixture.


©2017 This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

This article appeared in the August 2017 issue of International Research Journal of Computer Science (IRJCS).

The full-text of this article is available at



Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Publisher Citation

E. Aidan, C. Jennifer, and H. Coyle. Research report, IRJCS, Issue VIII. Forensic Science Department, Henry C. Lee College of Criminal Justice & Forensic Sciences, University of New Haven, United States, Forensic Science Department, Henry C. Lee College of Criminal Justice & Forensic Sciences, University of New Haven, United States, (August 2017)