When processing a trace DNA sample at the Netherlands Forensic Institute, an STR electropherogram can be created. An analyst uses this electropherogram and analysis software to read out peaks signifying DNA. After analysis, the DNA profile is used in the interpretation process, which can include the comparison to a reference DNA profile of a person of interest. The software that is currently being used for profile analysis is threshold-based and the process includes the intervention of trained analysts. To further automate (allelic) peak identification in STR electropherograms, as well as to increase efficiency and uniformity, neural networks were studied and applied. Previous work by Duncan Taylor and David Powers provided a proof of concept using a simple fully connected neural net. After reviewing literature, the U-net was selected to be used in this thesis. Training U-net on electropherograms proved successful and achieved a 95% accuracy on the per-pixel labels. However, translating the per-pixel output to alleles was more difficult than expected, so an upper bound on the score was calculated. The upper bound got close to an analyst’s performance and demonstrated the potential of this method.

In criminal investigations, individuals may be connected to illicit activities by linking their personal phone to an otherwise anonymous, crime-related phone. Several methods have been published that use cell tower registrations to differentiate between same-user and different-user scenarios for the two phones. However, criminals may deviate in movement patterns and phone usage from the test subjects on which the methods are developed and evaluated. Whether the proposed methods are robust to such different behavioral profiles is unclear. The scarcity of readily available datasets on criminals' movements and phone usage further complicates this issue. Lacking precise knowledge of the behavior of the population of interest, we propose a robustness analysis. Here, we present a tool for generating synthetic datasets, based on well-established models for the movement of individuals. We used the tool to generate data for a range of behavioral properties, encompassing variations in both underlying movement and phone usage. We evaluated three existing methods using our synthetic data. The first is a discriminatory approach that learns typical movement patterns and phone usage from a reference dataset. The second approach uses a model of cell tower behavior, making minimal assumptions on user behavior by choosing pairs of registrations close in time. The third is a generic statistical method for comparing event data. Additionally, we present a fourth method that combines the latter two, as conceptually, they use different aspects of the data. Our analysis reveals that the discriminatory method performs best in a baseline scenario but is most sensitive to behavioral deviations. The cell tower method shows the lowest baseline performance yet exhibits the strongest resilience to variations. The generic model appears intermediate in terms of performance and sensitivity. Given the importance of robustness in evaluating evidence, we recommend using the combined approach, which is both reliable and effective across our defined variations.