Enhanced sample preparation and data interpretation strategies using massively parallel sequencing for human identification in missing persons' and DVI casework

When the remains of victims from mass disasters, military conflicts, or missing persons’ cases are recovered, identification is the most important objective. The recovered unidentified remains may be intact, fragmented, comingled, highly decomposed, or skeletonized. The DNA within these tissues is often degraded, damaged, and/or contains inhibitory agents depending on the environment in which the remains were discovered. This project explores the use of traditional genotyping and newer DNA sequencing technologies for the identification of challenging human remains commonly recovered from mass disasters and missing persons’ cases. The results of this study will provide the forensic community with additional information on the comparative performance of massively parallel sequencing (MPS) chemistries and platforms with compromised samples, particularly highly inhibited samples. This study was comprised of four projects. First, two CE-based STR megaplex kits (GlobalFiler® PCR Amplification and Investigator® 24plex QS kits) were evaluated for their tolerance to PCR inhibitors (humic acid, melanin, hematin, collagen, calcium) and overall sensitivity of detection for high and low quantity (1 ng and 0.1 ng) DNA samples. The results suggested that the GlobalFiler® kit was more sensitive down to 7.8 pg of DNA while the Investigator® kit was more tolerant to all PCR inhibitors at both DNA concentrations. The GlobalFiler® kit produced more alleles, higher peak heights, and higher peak height ratios when determining sensitivity. Conversely, the Investigator® kit produced more alleles and balanced profiles for every inhibitor and inhibitor concentration than the GlobalFiler® kit. Second, two MPS chemistries and platforms (Ion AmpliSeq™ kit on the Ion PGM and the ForenSeq™ kit on the MiSeq FGx™) were evaluated side-by-side using the same inhibited DNA samples. The AmpliSeq™ and ForenSeq™ kits were found to be tolerant and susceptible to different common PCR inhibitors. The AmpliSeq™ chemistry demonstrated tolerance to collagen and calcium; however, it was highly susceptible to humic acid and hematin. Conversely, the ForenSeq™ kit showed extreme tolerance to hematin and calcium inhibitors but was greatly affected by melanin. The third study focused on determining the effectiveness of common DNA extraction methods to remove inhibitors from forensically relevant samples and their downstream compatibility with two MPS chemistries. Three substrates (blood, hair, and bone) were spiked with high concentrations of four inhibitors (humic acid, melanin, hematin, and calcium) and extracted using five DNA extraction methods (DNA IQ™, QIAamp® DNA Investigator, PrepFiler®, and two total demineralization protocols (bone only)). The results showed that all extraction methods were able to efficiently remove all PCR inhibitors with no sign of inhibition and provide sufficiently pure DNA extracts for sequencing. Although the amount of DNA recovered using the different extraction methods differed, the sequencing data indicated that none of the extraction methods negatively influenced the downstream sequencing performance on either MPS system. The fourth and final study reports the comparative performance of two MPS systems when sequencing challenging human skeletal remains. Thermally degraded, embalmed, cremated, burned, and decomposed bones and teeth (N = 24) were extracted using a total demineralization protocol and processed with two MPS chemistries and platforms in addition to traditional CE-based STR typing. The results demonstrated that CE-based STR profiling was still a valuable approach by providing at least a partial DNA profile for every sample, whereas MPS did fail to produce a profile in some instances. However, these MPS chemistries are still not fully optimized to tolerate such difficult samples and further optimization is warranted. Conversely, MPS has the capability to analyze more markers and multiple marker systems (STRs, SNPs, etc.) simultaneously. Therefore, even though some CE samples produced more complete profiles, the additional markers within MPS multiplexes may result in higher powers of discrimination for identification, and thereby provide results to assist with solving missing persons’, forensic, and DVI cases