Specific fluorescent signatures for body fluid identification using fluorescence spectroscopy.

Non-invasive, rapid, on-site detection and identification of body fluids is highly desired in forensic investigations. The use of fluorescence-based methods for body fluid identification, have so far remain relatively unexplored. As such, the fluorescent properties of semen, serum, urine, saliva and fingermarks over time were investigated, by means of fluorescence spectroscopy, to identify specific fluorescent signatures for body fluid identification. The samples were excited at 81 different excitation wavelengths ranging from 200 to 600 nm and for each excitation wavelength the emission was recorded between 220 and 700 nm. Subsequently, the total emitted fluorescence intensities of specific fluorescent signatures in the UV-visible range were summed and principal component analysis was performed to cluster the body fluids. Three combinations of four principal components allowed specific clustering of the body fluids, except for fingermarks. Blind testing showed that 71.4% of the unknown samples could be correctly identified. This pilot study shows that the fluorescent behavior of ageing body fluids can be used as a new non-invasive tool for body fluid identification, which can improve the current guidelines for the detection of body fluids in forensic practice and provide the robustness of methods that rely on fluorescence.

Improving the visualization of fingermarks using multi-target immunolabeling.

The development of fingermarks is an important step in visualizing ridge patterns for individualization purposes. Immunolabeling can be applied to fingermarks to selectively and sensitively detect antigens in fingermarks, and can be used as a developing method to visualize fingermarks. In this study we investigated single (the detection of one antigen) and multiple targeting approaches (the detection of multiple antigens simultaneously) to improve fingermark development. The detection of dermcidin, an antimicrobial peptide, was used as the gold standard to compare single and multi-target detection of keratins, albumin and/or dermcidin. Single detection of dermcidin and albumin mostly resulted in clear ridge details and/or pore detection, whereas the single keratin detection resulted in a poor visualization of the fingermarks. The multi-target approach in which both dermcidin and albumin were targeted, resulted in improved fingermark development compared to single dermcidin detection. Therefore, we recommend the use of multi-target detection consisting of anti-dermcidin and anti-albumin when using immunolabeling as fingermark development technique. Additionally, the optimized multi-target approach was tested as a pre- and post-development technique in combination with powder dusting and cyanoacrylate fuming. Immunolabeling has not been implemented yet in forensic case work, however we expect that immunolabeling can be used to redevelop poorly developed and/or smudged fingermarks in the nearby future. Currently, an ongoing pilot-study is being conducted in collaboration with the Dutch police.

Prediction of DNA concentration in fingermarks using autofluorescence properties.

During criminal investigations trace DNA samples, including fingermarks, are submitted to laboratories for short tandem repeat (STR) analysis. For most common STR analysis systems a minimum amount of input DNA is required. Upon intake by the forensic laboratory the DNA concentration is estimated using quantitative polymerase chain reaction (qPCR) analysis after which most fingermarks are excluded. To tackle the problem of unnecessary processing in the lab, our study aimed to develop a method, which is able to predict the DNA content in fingermarks directly at the crime scene. Upon excitation with a UV Crime-lite, fingermark residues have autofluorescent properties. We hypothesize that the intensity of the autofluorescence signal of the fingermark content correlates to the DNA concentration in fingermarks. In this study, 164 fingermarks were examined on their autofluorescence intensity when excited at 365nm, the number of nucleated cells, their DNA concentration and the completeness of the STR profiles. No significant correlation was observed between the DNA concentration in fingermarks and the autofluorescence signal, indicating that a high amount of autofluorescence, thus a high amount of biomaterial, does not necessarily guarantee a higher amount of DNA. In addition, the completeness of the STR profiles did not correlate to the autofluorescence signal of fingermarks. A moderate correlation was found between the predicted DNA quantity, based on the number of nucleated cells and the DNA quantity. In summary, the autofluorescence signal of fingermarks cannot directly be used as a guide to select fingermarks for DNA analysis directly at the crime scene. However, predicting the amount of DNA using a sensitive and specific DNA staining method can probably be used to estimate the DNA concentration in touch samples.