Development of a Mitochondrial Mixture Database and Interpretation Tool

As submitted by the proposer:

Mitochondrial DNA (mtDNA) typing is one of the most sensitive tests available for analyzing biological evidence. However, mtDNA typing is an underutilized method, with almost all work to date focusing on single-source samples (i.e., hairs, bones, teeth). One reason for this unmet potential is the chemistry of the assay; current mtDNA typing methods are based on Sanger sequencing and are labor intensive, costly, and, most importantly, not sufficiently quantitative for mixture interpretation and deconvolution.

This quantitative limitation resulted in mtDNA analysis rarely being applied to trace (or touch) DNA samples (i.e., samples that could be better analyzed by a more sensitive method). However, with the advent of massively parallel sequencing (MPS) the limitations of Sanger sequencing can be overcome. The increased throughput of MPS has two distinct advantages. First, while Sanger-based techniques previously targeted the ~600bp hypervariable region, MPS can quickly and easily assay the entirety of the 16,569bp mitochondrial genome (mtgenome), allowing for greater discriminatory power. Second, MPS is easily parallelized, such that many forensic and reference samples can be analyzed simultaneously.

The sensitivity of MPS mtDNA typing is such that input DNA less than or equal to 10 pg can yield whole mtgenome results. The sensitivity of detection, dynamic range, and quantitative nature of MPS combined with the high copy number of mtgenomes per cell and access of a small size amplicon multiplex that spans the entire mtgenome now make mtDNA a potentially useful marker for a wider variety of types of forensic biological evidence, including touch DNA. To exploit the potential of MPS with mtDNA typing it is imperative to be able to interpret mixture evidence. The general principles of mixture interpretation are well-understood and the MPS methodology is sufficiently robust and quantitative.

This proposal addresses interpretation strategies, software development, and empirical verification of mtgenome mixture evidence. The software will facilitate mixture interpretation, be user friendly, freely available, and easily configurable. Furthermore, it will house greater than or equal to 10,000 whole mtgenome haplotypes for generation of synthetic mixtures, and it will perform searches of single-source and mixture profiles and calculate appropriate statistics for mixtures. It also will support user input features, such as thresholding on coverage and selecting the number of contributors.

Single-source data are available to define basic performance data and mixture profiles can be readily generated. This universal tool will bring to the forefront the capability of mixture analyses of mtgenome data and expand the capabilities of forensic scientists to analyze challenged biological evidence.

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