Current assessment of whether a forensic evidence item should be submitted for STR profiling is largely based on the personal experience of the Crime Scene Investigator (CSI) and the submissions policy of the law enforcement authority involved. While there are chemical tests that can infer the presence of DNA through the detection of biological stains, the process remains mostly subjective and leads to many samples being submitted that give no profile or not being submitted although DNA is present. The ParaDNA(®) Screening System was developed to address this issue. It consists of a sampling device, pre-loaded reaction plates and detection instrument. The test uses direct PCR with fluorescent HyBeacon™ detection of PCR amplicons to identify the presence and relative amount of DNA on an evidence item and also provides a gender identification result in approximately 75 minutes. This simple-to-use design allows objective data to be acquired by both DNA analyst and non-specialist personnel, to enable a more informed submission decision to be made. The developmental validation study described here tested the sensitivity, reproducibility, accuracy, inhibitor tolerance, and performance of the ParaDNA Screening System on a range of mock evidence items. The data collected demonstrates that the ParaDNA Screening System identifies the presence of DNA on a variety of evidence items including blood, saliva and touch DNA items.

No abstract

A central tenet of probabilistic programming is that a model is specified exactly once in a canonical representation which is usable by inference algorithms. We describe JointDistributions, a family of declarative representations of directed graphical models in TensorFlow Probability.

No abstract