Probabilistic Genotyping for Microhaplotypes

Microhaplotypes (MHs) are an emerging forensic DNA marker characterized by sets of single nucleotide polymorphisms (SNPs) within a short distance of each other displaying multiple allelic combinations. Although less polymorphic than short tandem repeat polymorphisms (STRs) they have some advantages such as alleles all of the same size within a locus, absence of stutter artifacts, and mutation rates is significantly lower than that of STRs. Several MH-multiplex panels have been reported in the past two years, including the 74-locus panel developed in our laboratory. Casework implementation of such large panels is only feasible if paired with probabilistic genotyping (PG) as manual deconvolution of complex mixtures would be excessively time consuming and not compatible with conventional forensic DNA laboratory operations. To evaluate the feasibility of this project LRmix Studio and DNA-View Mixture Solution were preliminarily adapted to processing MH data from 74 loci analyzed on the Ion SS massively parallel sequencing (MPS) platform. Relative fluorescence unit (RFU) values were replaced by allele-sequence coverage and tested on a set of DNA mixtures. Results show that MHs data can be interpreted using PG, yet more work is necessary to develop sound stochastic modeling of allele coverage ratio (ACR) and drop out for effective deconvolution. The goal of this project is to (1) adapt and thoroughly test three PG platforms (LRmix Studio, EuroForMix, and DNA-View Mixture Solution) for the use multiplex microhaplotype data for DNA mixture interpretation, and (2) generate a publicly available data repository (mixtures and references) that can be used, by developers and users, to adapt other PG software to intake MH data. In the first phase of the project a set of dilutions will be performed to evaluate stochastic variation at low input DNA levels. In parallel, a series of simple and complex mixtures, ranging from two (2) to five (5) contributors in multiple donor-ratio combinations, will be prepared and analyzed integrating the data already generated as part of a current effort (2017-DN-BX-0164) for a total of 100 mixtures. These will also be STR typed (both CE and MPS). The data will be made publicly available to the forensic community. The second phase of the project will be adapting the previously mentioned PG software and evaluating its performance with MHs vs STRs. Results will be published on peer reviewed journals. This project represents the next logical phase for making this novel promising forensic marker one step closer to casework implementation. 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