Quantifying the accuracy of low-quality DNA sample analysis from genotyping to genealogical searching and integration as a bioinformatic pipeline

Interest in investigative genetic geneaology (IGG) in the forensic science field is rapidly increasing as it has enabled investigators to solve murder and missing person cases that had long gone cold. In these cases, short fragments of DNA were sequenced to genotype SNPs and uploaded to databases such as GEDMatch for genealogical tracing. However, decisions during both the genotyping and genealogical searching steps are currently ad hoc, leading to both underpowered analyses, as well as overconfidence in tenuous results. Our objective is to utilize simulations to characterize the accuracy and limitations of both genotyping of degraded DNA and subsequent genealogical searching, developing best practices recommendations for varying sample qualities. By performing simulations from a range of populations we will also quantify the impacts of reference bias and estimations of genetic relatives for individuals from populations underrepresented in current reference databases. After we have completed the proposed work, the field will have an increased understanding of the variability and limitations of the different methodologies used in genotyping and genealogical software and their impacts on the accuracy of the identification of familial relationships, particularly for individuals underrepresented in the population databases typically used in analyses.

Note: This project contains a research and/or development component, as defined in applicable law," and complies with Part 200 Uniform Requirements - 2 CFR  200.210(a)(14). CA/NCF