Developing SPLiT-Seq: A Massively Parallel Sequencing Method to Remove PCR Artifacts for Forensic STR Genotyping

As submitted by the proposer:

The use of short-tandem repeats (STR) in forensic DNA analysis has been successful in most situations. However, this technology has several important limitations, including difficulty in identifying individuals occurring at a minor allele frequency (MAF) of less than 10 percent, mixtures with less than two people, and damaged DNA samples. The advent of massively parallel sequencing (MPS) has promised to revolutionize forensic DNA analysis due to increased sensitivity in detecting lower MAFs. Unfortunately, MPS relies on PCR during sample preparation, which, introduces a significant amount of artifacts, such as PCR stutter, and limiting its utility for forensic analysis.

To overcome the high error-rate of MPS and facilitate the genotyping of damaged DNA samples and heterogeneous DNA mixtures, the researchers previously developed a highly-accurate MPS methodology called "duplex sequencing," (DS) which delivers an unprecedented >10,000-fold improvement in accuracy compared to conventional MPS by allowing two strands of a DNA molecule to be compared to one another. However the DS protocol is slow, cumbersome, and requires >100ng of DNA.

To address this issue, the researchers recently developed a modified form of DS called SPLiT-Seq (separated PCRs of linked templates for sequencing), which relies solely on PCR for enrichment, but maintains the accuracy of DS while using <10ng of starting DNA.

The proposed project will develop SPLiT-Seq for forensic DNA analysis and is split into three specific aims: 1) Develop a SPLiT-Seq protocol for the CODIS20 loci and evaluation of accuracy, precision, sensitivity, and specificity; 2) Validation of SPLiT-Seq on a population of single-source samples from the Personal Genome Project; and 3) Validation of SPLiT-Seq for genotyping damaged DNA and DNA mixtures. To facilitate these goals and begin introducing our methods to forensics community, we have entered into a collaboration with the Washington State Patrol Crime Lab Division to work with us in validating SPLiT-Seq against current best practices.

At the conclusion of the project, the researchers expect to have developed a robust sample preparation protocol for SPLiT-Seq that has been validated against both current capillary electrophoresis and MPS genotyping methods. In addition, the researchers will have evaluated its performance on damaged DNA and DNA mixtures. To help disseminate this work to the appropriate audience, the researchers anticipate publishing two-to-three peer reviewed articles, and presenting their work at conferences and seminars.

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