Figure 7 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Probability density distributions for the log difference between all compound LRs in the study and their corresponding simple LR products (a), as well as all conditioned LRs in the study and their unconditioned counterparts (b). An overlay of the two distributions (c) highlights their similarity.
Source publication
In cases where multiple questioned individuals are separately supported as contributors to a mixed DNA profile, guidance documents recommend performing a comparison to see if there is support for their joint contribution. Anecdotal observations suggest the summed log of the individual likelihood ratios (LR), termed the simple LR product, should be...
Contexts in source publication
Context 1
... not every compound LR was higher than its simple LR counterpart, our expectations about compound LR magnitude are most straightforwardly defined in terms of upper and lower bounds. The distribution of log(LR) differences between all compound LRs in this study and their corresponding simple LR products ranged from a minimum of ~−2.7 to a maximum of ~28.3, with the probability density peaking at approximately 0.5 (see Figure 7a). Figures 1a,c and 2a,c, for the 2-and 3-person mixtures, are the most striking in this regard; mixtures with contributors in the same proportions have regular, distinct datapoint groupings, and the trends for the mixtures with more inherent ambiguity (e.g., 1:1, 1:1:1 and 100:100:4) are translated further upward than the trends for the more easily resolved mixture proportions. ...
Context 2
... not every compound LR was higher than its simple LR counterpart, our expectations about compound LR magnitude are most straightforwardly defined in terms of upper and lower bounds. The distribution of log(LR) differences between all compound LRs in this study and their corresponding simple LR products ranged from a minimum of ~−2.7 to a maximum of ~28.3, with the probability density peaking at approximately 0.5 (see Figure 7a). ...
Context 3
... in a similar fashion to the compound/LR product plots, the conditioned/unconditioned LR plots show that conditioning does not always lead to a higher LR. The distribution of log(LR) differences between conditioned LRs and their corresponding unconditioned counterparts ranged from a minimum of ~−3.2 to a maximum of ~27.7, with the probability density again peaking at approximately 0.5 (see Figure 7b). The similarity of the log(LR) difference distributions for the compound/simple and conditioned/unconditioned comparisons (see Figure 7c) points to parallel mechanisms of action. ...
Context 4
... distribution of log(LR) differences between conditioned LRs and their corresponding unconditioned counterparts ranged from a minimum of ~−3.2 to a maximum of ~27.7, with the probability density again peaking at approximately 0.5 (see Figure 7b). The similarity of the log(LR) difference distributions for the compound/simple and conditioned/unconditioned comparisons (see Figure 7c) points to parallel mechanisms of action. In fact, because the weights produced by STRmix for the various genotype combinations can be characterized as either unconditional or conditional probabilities, some connections can be drawn between the effects of compound LRs and conditioned LRs. ...
Context 5
... in a similar fashion to the compound/LR product plots, the conditioned/unconditioned LR plots show that conditioning does not always lead to a higher LR. The distribution of log(LR) differences between conditioned LRs and their corresponding unconditioned counterparts ranged from a minimum of ~−3.2 to a maximum of ~27.7, with the probability density again peaking at approximately 0.5 (see Figure 7b). ...
Context 6
... similarity of the log(LR) difference distributions for the compound/simple and conditioned/unconditioned comparisons (see Figure 7c) points to parallel mechanisms of action. In fact, because the weights produced by STRmix for the various genotype combinations can be characterized as either unconditional or conditional probabilities, some connections can be drawn between the effects of compound LRs and conditioned LRs. ...
Similar publications
After the onset of COVID-19 pandemic, a sharp surge in the usage of the face-masks throughout the globe has been observed. Pre-experiment survey of 252 individuals indicated a higher use of cotton-make masks (41%), followed by N-95 make (31%), and surgical disposable masks (26%). It was also further revealed that a higher fraction of individuals we...
Zusammenfassung
Die biostatistische Bewertung DNA-analytischer Befunde unterstützt Gerichte bei der Einschätzung des Beweiswertes einer Spur. In der Praxis werden dabei zunehmend Spuren mit minimaler DNA-Menge und möglichen „Drop-in“- und „Drop-out“-Ereignissen sowie komplexe Mischspuren analysiert. Solche Spuren sind mit einer klassischen „binären...
Citations
... The LR is calculated based on the evidence and potential information from the population. In the case of DNA, this is the frequency of polymorphisms obtained from population studies [62][63][64][65]. Similarly, knowledge of the frequency with which different traits, minutiae, deltas, ridge flows, and so forth occur in different human populations, as well as their topological, finger and sexual variations, is a relevant aspect that would improve the identification process and should be taken into account. ...
... In this case, it is prudent to test whether these POI could explain the profile when considered together. This could be undertaken using a compound proposition pair, defined as one where more than one POI within H p is replaced with unknown donors in H a ( [8], hereafter the ASB (American Standards Board) draft standard and see also [9,10] Although this proposition pair is highly effective in assessing whether both POI could be donors together, reported without the simple LRs for each individual, it can appear to greatly overstate the weight against a POI who gives a small inclusionary or uninformative LR when considered individually but who is carried in the compound LR by the much stronger other donors to the mixture. ...
... For example, conditioning on only one or two known contributors within a four-person mixture. These partial conditioned LRs are not calculated within this paper but are explored by Duke et al. [9] (see, for example, the study's Table 4). ...
... In these cases, we recommend the use of extended MCMC accepts within the interpretation. This allowed for more time to explore the sample space and was also a finding of Duke et al. [9]. ...
Simple propositions are defined as those with one POI and the remaining contributors unknown under Hp and all unknown contributors under Ha. Conditional propositions are defined as those with one POI, one or more assumed contributors, and the remaining contributors (if any) unknown under Hp, and the assumed contributor(s) and N unknown contributors under Ha. In this study, compound propositions are those with multiple POI and the remaining contributors unknown under Hp and all unknown contributors under Ha. We study the performance of these three proposition sets on thirty-two samples (two laboratories × four NOCs × four mixtures) consisting of four mixtures, each with N = 2, N = 3, N = 4, and N = 5 contributors using the probabilistic genotyping software, STRmix™. In this study, it was found that conditional propositions have a much higher ability to differentiate true from false donors than simple propositions. Compound propositions can misstate the weight of evidence given the propositions strongly in either direction.
... The −1 repeat stutter variance was set to be inversely proportional to the observed height of the parent allele; all other stutter variances were set to be inversely proportional to the expected stutter peak height (personal communication with STRmix support staff, 1 September 2020). MCMC accepts per each of the 8 chains were set to 10,000 burn-in/50,000 post burn-in for single-source, 2-person mixtures and 3-person mixtures, and to 200,000 burn-in/1,000,000 post burn-in for 4-person mixtures [12]. The Gelman-Rubin autocontinue option was activated for all runs, and was set to add 10,000 post burn-in accepts to any MCMC initially producing a Gelman-Rubin convergence diagnostic in excess of 1.2. ...
Distributions of the variance parameter values developed during the validation process. Comparisons of these prior distributions to the run-specific average are one measure used by analysts to assess the reliability of a STRmix deconvolution. This study examined the behavior of three different STRmix variance parameters under standard amplification and interpretation conditions, as well as under a variety of challenging conditions, with the goal of making comparisons to the prior distributions more practical and meaningful. Using information found in STRmix v2.8 Interpretation Reports, we plotted the log10 of each variance parameter against the log10 of the template amount of the highest-level contributor (Tc) for a large set of mixture data amplified under standard conditions. We observed nonlinear trends in these plots, which we regressed to fourth-order polynomials, and used the regression data to establish typical ranges for the variance parameters over the Tc range. We then compared the typical variance parameter ranges to log10(variance parameter) v log10(Tc) plots for mixtures amplified and interpreted under a variety of challenging conditions. We observed several distinct patterns to variance parameter shifts in the challenged data interpretations in comparison to the unchallenged data interpretations, as well as distinct shifts in the unchallenged variance parameters away from their prior gamma distribution modes over specific ranges of Tc. These findings suggest that employing empirically determined working ranges for variance parameters may be an improved means of detecting whether aberrations in the interpretation were meaningful enough to trigger greater scrutiny of the electropherogram and genotype interpretation.
