This suggests that the ParaDNA Sample Collector recovers

a small proportion of the available DNA and any impact that the ParaDNA Sample Collector has on the level of subsequently available DNA is masked by the overall variability in yield caused by variation in sample preparation, swabbing efficiency and DNA recovery. STR Autophagy Compound Library typing was performed on all samples that gave a quantification result of ≤ 50 pg/μl. Using the amplification of 14 or more alleles as a benchmark indicator that the SGMPlus profile was ‘usable’, samples were categorised as either True Positives, True Negatives, False Positives, or False Negatives (Table 1). Samples that yielded more than 50 pg/μl of DNA were assumed suitable to provide a full SGM Plus profile. The data displayed in Table 1 indicate that blood and saliva consistently gave accurate results, while the touch DNA samples contained some instances where the ParaDNA and laboratory testing gave differing results. The STR profiling success rate of samples is known to vary [12], with touch DNA samples being amongst the poorest sample type submitted for STR profiling [14]. In this study the percentage of touch DNA samples (latex gloves, tools, fingerprints) that gave an STR profile of ≥ 14 alleles was 51% (42/83 samples). If the ParaDNA AZD5363 solubility dmso System had been used to identify which samples to preferentially submit for STR profiling the success rate of the submitted samples would

have been 82% (28/34 samples). While this represents a reduction in the number of successful profiles obtained from this group of 83 samples

(42 with no ParaDNA vs 28 with ParaDNA) it also represents a potential cost saving Acetophenone from the samples that were not submitted. This cost saving will allow a forensic service provider to screen and submit additional evidence items from other groups and thereby improve their overall success rate. It is not possible to assess whether the false negative rate presented in Table 1 obtained after using the ParaDNA Screening System is higher or lower than that achieved based on a traditional submissions approach as the identification of false negatives is only possible if there is a method to identify the false negatives. In practice, any item not currently submitted for STR profiling which would have given a full profile if submitted could be treated as equivalent to a false negative. Using the binary classification test to describe the proportion of true positives (sensitivity) and true negatives (specificity) [22] across all sample types (blood, saliva, and touch DNA) the system had a sensitivity of 86% and a specificity of 93%. The data presented above suggest that the ParaDNA System is capable of detection of DNA at low levels. The sensitivity and accuracy of the gender identification call in the ParaDNA assay are dependent on results from a single tube while the DNA Detection Score is summed from all four tubes.