A new era in forensic intelligence : SNPs and the inference of biogeographical ancestry

University of Technology, Sydney. Faculty of Science.NO FULL TEXT AVAILABLE. This thesis contains 3rd party copyright material. The hardcopy may be available for consultation at the UTS Library.NO FULL TEXT AVAILABLE. This thesis contains 3rd party copyright material. ----- The past few decades have witnessed continuous growth and development in the field of forensic DNA analysis. This remarkable evolution has been the result of a combination of vast technological advances in DNA analytical techniques, an increased knowledge and understanding of human genetics, and the identification, characterisation and cataloguing of DNA polymorphisms of medical, population genetic and forensic importance. While the use of STRs in forensic DNA analysis has been an enormous success, applications of forensic DNA profiling have focused, almost ubiquitously, on resolving the identity of the donor of a biological evidence sample. However, the potential to extend the analysis of such samples to retrieve information relating to the biogeographical ancestry, or physical characteristics, of the donors has become apparent. SNPs have been widely investigated as markers of biogeographical ancestry and are good candidate markers for this purpose due to their low mutation rate. In addition to being located in non-coding regions of the genome, SNPs are also located in regulatory and coding regions which confer potential for a role for SNPs in human functional variation, such as phenotypic variation. The development of SNP-based forensic DNA assays for the inference of biogeographical ancestry and physical characteristics is limited but has revealed enormous potential in forensic intelligence. DNA-based intelligence tools could be used to provide inferred descriptions of offenders from biological evidence samples and could be utilised in the primary (and critical) stages of a forensic investigation. Such information could be highly advantageous in narrowing a pool of potential suspects and would enable a more efficient use of valuable police and forensic resources. There is also obvious potential for such techniques in large-scale disaster victim identification cases. This thesis details the development of a six (6plex) and sixteen (16plex) autosomal SNP locus multiplex assay for the inference of biogeographical ancestry. The initial stage (Phase I) of the project involved the development (and optimisation) of a 6plex assay to examine its utility in inferring biogeographical ancestry of individuals from five major populations of interest in the Australian community; African, Asian, Caucasian, Middle Eastern and Sub-Continental Asian populations. The SNP loci selected for inclusion in the 6plex assay were those that demonstrated population specific (skewed) allele frequencies as determined by allele and genotype frequency population data collated from the literature. The 6plex assay was used to genotype 156 population samples. The genotype data was used to assess population genetic parameters and to infer biogeographical ancestry (using the Linear Discriminate Function and Maximum Likelihood Estimates). This analysis revealed a high degree of accuracy for the inference of African individuals (100%), an intermediate degree of accuracy for Asian (65%) and Caucasians (67%) and a lower degree for Middle Eastern (44%) and Sub-Continental Asians (27%). Phase II of this project involved the expansion of the 6plex assay to increase the inference accuracy from population samples which revealed intermediate or low accuracies. An additional ten SNP loci were selected and a sixteen locus (16plex) SNP multiplex assay was developed and optimised. Pigmentation-related SNPs were also investigated for their utility in eye and hair colour inferences. A total of 242 population samples (divided into six major population groups Sub-Saharan African, North African, Asian, Sub-Continental Asian, Middle Eastern and Caucasian) were analysed using the 16plex assay. Inference of biogeographical ancestry was conducted using Maximum Likelihood Estimates and revealed a high degree of accuracy for Asian (90.1%), Caucasian (91.1%), Sub-Saharan African (100.0%) and North African (90.0%) samples. The inference of Middle Eastern (60.0%) and Sub-Continental Asian (78.8%) samples was also significantly improved. Sensitivity studies of the l6plex assay revealed complete SNP profiles were achievable using 1.25ng of DNA template. A preliminary casework validation revealed the applicability of the 16plex assay to routine forensic casework samples. This study has demonstrated the utility of SNPs to infer biogeographical ancestry and is the result of research not previously conducted in Australia. Further development of the l6plex assay, and increased understanding of SNPs, is required to assess the ultimate relevance and utility of SNPs in this capacity