Developments in forensic DNA analysis

Abstract

The analysis of DNA from biological evidence recovered in the course of criminal investigations can provide very powerful evidence when a recovered profile matches one found on a DNA database or generated from a suspect. However, when no profile match is found, when the amount of DNA in a sample is too low, or the DNA too degraded to be analysed, traditional STR profiling may be of limited value. The rapidly expanding field of forensic genetics has introduced various novel methodologies that enable the analysis of challenging forensic samples, and that can generate intelligence about the donor of a biological sample. This article reviews some of the most important recent advances in the field, including the application of massively parallel sequencing to the analysis of STRs and other marker types, advancements in DNA mixture interpretation, particularly the use of probabilistic genotyping methods, the profiling of different RNA types for the identification of body fluids, the interrogation of SNP markers for predicting forensically relevant phenotypes, epigenetics and the analysis of DNA methylation to determine tissue type and estimate age, and the emerging field of forensic genetic genealogy. A key challenge will be for researchers to consider carefully how these innovations can be implemented into forensic practice to ensure their potential benefits are maximised.

Keywords: DNA profiling; forensic biology; forensic genetics.

Figure 1.. Number of publications returned in a search of www.scopus.com for ‘forensic AND DNA’ in the title, abstract or keywords, for the years 1990 to 2020.

Figure 2.. Nucleotide sequence of three alleles detected at the vWA locus.

All three of these alleles would be classified as 14 alleles on the basis of their length but determining their nucleotide sequence allows them to be discriminated.

Figure 3.. Electropherogram showing part of a mixed DNA profile at four STR loci.

Across a whole profile, interpretation of the varying number of peaks and peak heights can become very complex, even with a small number of contributors.