The ability to obtain deoxyribonucleic acid (DNA) profiles is generally considered a powerful tool when examining evidence associated with a crime scene. However, variability in peak heights associated with short tandem repeats (STR) signal complicates DNA interpretation; particularly, low-template complex mixtures, which are regularly encountered during evidentiary analysis. In order to elucidate the sources that cause peak height variability a dynamic model, which simulates; 1) the serial dilution process; 2) polymerase chain reaction (PCR); and 3) capillary electrophoresis (CE) was built and used to generate simulated DNA evidentiary profiles.
In order to develop the dynamic model, PCR efficiencies were characterized. This was accomplished using empirical quantitative polymerase chain reaction (qPCR) data. Specifically, the ratios of fluorescent readings of two consecutive cycles were evaluated. It was observed that the efficiency fluctuated at early cycles; stabilized during the middle cycles; and plateaued during later cycles. The relationship between the change in efficiency and the concentration of amplicons was modeled as an exponential function. Subsequently, this exponential relationship was incorporated into the dynamic model as a part of the PCR module.
Using the dynamic laboratory model, the effect of serially diluting a concentrated DNA extract to a low-template concentration was assessed in an effort to determine whether serially diluted samples are a good representation of evidence samples which contain low copy number of cells. To accomplish this, peak height variances and the frequency of drop-out between serially and non-serially diluted samples were compared. The results showed that diluting the sample had a substantial influence on allelic drop-out. However, the distributions of the observed peak heights did not consistently change; though, changes in peak height distributions became more pronounced with samples at lower targets. The peak height equivalency (PHE) was also used to aid in the determination of the effect of serial dilutions on reproducibility. There was not a major change in PHE between serially and non-serially diluted samples.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/16242 |
| Date | 08 April 2016 |
| Creators | Hu, Cheng-Tsung |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Rights | Attribution 4.0 International, http://creativecommons.org/licenses/by/4.0/ |
Page generated in 0.0173 seconds