Forensic Microbial Analysis of Biological Fluids for Identification and Time since Deposition Estimation: A Systematic Literature Review

Abstract

Microbial analysis is a new technique that is being developed in forensic science for biological fluids and determining time since deposition (TsD). This is a systematic literature review according to PRISMA guidelines in order to evaluate the current state of the microbial metagenomics in forensics. A total of 34 investigations of the bacteriological and/or fungal profile of blood, semen, saliva, vaginal fluid and urine were identified. The review highlights the transition from DNA analysis of humans to using the microbial taxa as stable biomarkers.

Microbial profiling can be used to assist in the identification of a biological fluid as a forensic tool with generally high classification performance, but is highly dependent on the fluid, its matrix and the design. In the literature surveyed, the accuracy of the identification general ranged between 79% and 94.6%, being especially high for classes of saliva, vaginal fluid, feces, semen and blood related classes, when using panels of markers or machine-learning classifiers. Conventional presumptive tests are still hampered by a lack of specificity, with some microbial signatures able to persist even when human RNA is degraded and others able to be used to support a STR typing from the same extracts as part of workups. It is also evident that in complex discriminatory tasks, a single markers set is not enough, and a multiple panel of markers, a probabilistic interpretation and classification models trained with a wide range of forensic-like samples are necessary for full discrimination.

It is important to take into consideration the temporal and environmental effects because of the possibility of having site-specific profiles that do not mean the same as those found in the environment, which environmental conditions become more "there" over time or after exposure in the outdoors. In general, the literature suggests that microbial evidence can be best used as a complementary rather than independent forensic technique and that vaginal fluid, saliva, semen, faces and mixed source are most likely to also be seen as meaningful evidence for this forensic technique, although the latter is a challenge due to the presence of more than one source in the evidence; inter-individual variation and incomplete standardization of the evidence are significant obstacles to the routine deployment of this technique in the casework.

The results suggest high throughput sequencing (such as 16S rRNA, MPS) could be very useful for the identification of fluids, however, estimation of TsD is still problematic because of variability (from individual to individual) and environmental influences. In the case of menstrual blood, however, specific models have low Mean Absolute Error (MAE) values of 1.54 days. This report summarizes the current understanding of microbial diversity as well as the effectiveness of using the biomarkers and the result of degradation on the forensic value.