Production of high-fidelity electropherograms results in improved and consistent DNA interpretation: Standardizing the forensic validation process

Abstract

Samples containing low-copy numbers of DNA are routinely encountered in casework. The signal acquired from these sample types can be difficult to interpret as they do not always contain all of the genotypic information from each contributor, where the loss of genetic information is associated with sampling and detection effects. The present work focuses on developing a validation scheme to aid in mitigating the effects of the latter. We establish a scheme designed to simultaneously improve signal resolution and detection rates without costly large-scale experimental validation studies by applying a combined simulation and experimental based approach. Specifically, we parameterize an in silico DNA pipeline with experimental data acquired from the laboratory and use this to evaluate multifarious scenarios in a cost-effective manner. Metrics such as signal-to-noise resolution, false positive and false negative signal detection rates are used to select tenable laboratory parameters that result in high-fidelity signal in the single-copy regime. We demonstrate that the metrics acquired from simulation are consistent with experimental data obtained from two capillary electrophoresis platforms and various injection parameters. Once good resolution is obtained, analytical thresholds can be determined using detection error tradeoff analysis, if necessary. Decreasing the limit of detection of the forensic process to one copy of DNA is a powerful mechanism by which to increase the information content on minor components of a mixture, which is particularly important for probabilistic system inference. If the forensic pipeline is engineered such that high-fidelity electropherogram signal is obtained, then the likelihood ratio (LR) of a true contributor increases and the probability that the LR of a randomly chosen person is greater than one decreases. This is, potentially, the first step towards standardization of the analytical pipeline across operational laboratories.

Four model variants within a continuous forensic DNA mixture interpretation framework: Effects on evidential inference and reporting

“…To calculate an approximation of the LR and the p -value of the LR, CEESIt samples 1 billion (10 9 ) genotypes r i from the set R 1 \{ s }. The LR is calculated as follows: where M = 10 9 [ 37 ].…”

A large-scale validation of NOCIt’s a posteriori probability of the number of contributors and its integration into forensic interpretation pipelines

“…Within the context of this study, the nominal NOC is taken to be the TrueNOC as it is reasonable to assume that each contributor's DNA is represented in the signal since it has previously been demonstrated that for this dataset: 1) the RFU signal from a single amplifiable molecule of DNA is fully resolved from noise at these laboratory and AT conditions [20]; 2) the smallest minor contributor of any mixture sample did not fall below 0.016 ng, which is approximately two copies of DNA (NB: some single-source samples were amplified at 0.008 ng); and 3) all 25 s GlobalFiler™ single-source PROVEDIt samples amplified at 0.016 ng or higher rendered RFU signal at a minimum of 12 known allele locations, regardless of the laboratory or sample condition [8]. Though it is likely that some signal at these locations may be attributed to noise, it is unlikely that all 12 can.…”

“…Thus, the dropout rate for the smallest minor in the first bin is expected to range between 28 and 6%. The second bin contains samples with minor contributors that have expected drop-out rates between 0.3 and 6%, whereas the remaining bins all have drop-out rates less than 0.3 % [20] when neither the DNA is degraded or PCR is inhibited. Thus, the bin ranges represent minor contributor quantities exhibiting considerable, some, low and extremely low allelic drop-out due to sampling.…”

“…The mixtures were generated under controlled laboratory conditions and so the actual number of contributors, i.e., TrueNOC, for each sample is known. These 815 samples were specifically selected to test the biological models because they represent mixtures that: 1) contain information from a kit containing the extended CODIS loci; 2) consist of 1-to 5-contributors, some of which were subjected to conditions that compromise the integrity of the PCR; and 3) were run using an optimized forensic DNA pipeline engineered to produce allele signal that is well resolved from noise when in the single-copy regime [20].…”

“…Since a decrease in the AT will necessarily increase the chance of detecting instrument and PCR artifacts -potentially increasing the burden associated with electropherogram review -we have implemented a module, named CleanIt, that automatically filters minus A (i.e., incomplete adenylation of newly polymerized DNA fragments), pull-up and raised baseline, using parameters provided by the laboratory. Though the application of an AT is unnecessary for the full interpretation pipeline, it remains an option as seemingly contradictory findings on the effects of ATs on LRs have been presented by the authors of [19], who demonstrated that for the 72 samples used in their study, a higher AT improved LR results for two probabilistic programs, and the authors of [20], who demonstrated that optimized ATs based on bioanalytical principles improve and stabilize inference outcomes across laboratory systems.…”