Ultrasensitive sequencing of STR markers utilizing unique molecular identifiers and the SiMSen-Seq method.

Massively parallel sequencing (MPS) is increasingly applied in forensic short tandem repeat (STR) analysis. The presence of stutter artefacts and other PCR or sequencing errors in the MPS-STR data partly limits the detection of low DNA amounts, e.g., in complex mixtures. Unique molecular identifiers (UMIs) have been applied in several scientific fields to reduce noise in sequencing. UMIs consist of a stretch of random nucleotides, a unique barcode for each starting DNA molecule, that is incorporated in the DNA template using either ligation or PCR. The barcode is used to generate consensus reads, thus removing errors. The SiMSen-Seq (Simple, multiplexed, PCR-based barcoding of DNA for sensitive mutation detection using sequencing) method relies on PCR-based introduction of UMIs and includes a sophisticated hairpin design to reduce unspecific primer binding as well as PCR protocol adjustments to further optimize the reaction. In this study, SiMSen-Seq is applied to develop a proof-of-concept seven STR multiplex for MPS library preparation and an associated bioinformatics pipeline. Additionally, machine learning (ML) models were evaluated to further improve UMI allele calling. Overall, the seven STR multiplex resulted in complete detection and concordant alleles for 47 single-source samples at 1 ng input DNA as well as for low-template samples at 62.5 pg input DNA. For twelve challenging mixtures with minor contributions of 10 pg to 150 pg and ratios of 1-15% relative to the major donor, 99.2% of the expected alleles were detected by applying the UMIs in combination with an ML filter. The main impact of UMIs was a substantially lowered number of artefacts as well as reduced stutter ratios, which were generally below 5% of the parental allele. In conclusion, UMI-based STR sequencing opens new means for improved analysis of challenging crime scene samples including complex mixtures.

US Population Data for 94 Identity-Informative SNP Loci.

The US National Institute of Standards and Technology (NIST) analyzed a set of 1036 samples representing four major US population groups (African American, Asian American, Caucasian, and Hispanic) with 94 single nucleotide polymorphisms (SNPs) used for individual identification (iiSNPs). The compact size of iiSNP amplicons compared to short tandem repeat (STR) markers increases the likelihood of successful amplification with degraded DNA samples. Allele frequencies and relevant forensic statistics were calculated for each population group as well as the aggregate population sample. Examination of sequence data in the regions flanking the targeted SNPs identified additional variants, which can be combined with the target SNPs to form microhaplotypes (multiple phased SNPs within a short-read sequence). Comparison of iiSNP performance with and without flanking SNP variation identified four amplicons containing microhaplotypes with observed heterozygosity increases of greater than 15% over the targeted SNP alone. For this set of 1036 samples, comparison of average match probabilities from iiSNPs with the 20 CODIS core STR markers yielded an estimate of 1.7 × 10-38 for iiSNPs (assuming independence between all 94 SNPs), which was four orders of magnitude lower (more discriminating) than STRs where internal sequence variation was considered, and 10 orders of magnitude lower than STRs using established capillary electrophoresis length-based genotypes.

Massively parallel sequencing data of 31 autosomal STR loci obtained using the Precision ID GlobalFiler NGS STR Panel v2 for 82 Japanese population samples.

Allele frequencies for 31 autosomal short tandem repeat (STR) loci (CSF1PO, D10S1248, D12ATA63, D12S391, D13S317, D14S1434, D16S539, D18S51, D19S433, D1S1656, D1S1677, D21S11, D22S1045, D2S1338, D2S1776, D2S441, D3S1358, D3S4529, D4S2408, D5S2800, D5S818, D6S1043, D6S474, D7S820, D8S1179, FGA, Penta D, Penta E, TH01, TPOX, and vWA) were obtained using Precision ID GlobalFiler NGS STR Panel v2 from 82 unrelated individuals sampled from the Japanese population. Autosomal STR alleles designated by NGS and conventional capillary electrophoresis were found to be concordant except at D2S441 allele 9.

Platinum-Quality Mitogenome Haplotypes from United States Populations.

A total of 1327 platinum-quality mitochondrial DNA haplotypes from United States (U.S.) populations were generated using a robust, semi-automated next-generation sequencing (NGS) workflow with rigorous quality control (QC). The laboratory workflow involved long-range PCR to minimize the co-amplification of nuclear mitochondrial DNA segments (NUMTs), PCR-free library preparation to reduce amplification bias, and high-coverage Illumina MiSeq sequencing to produce an average per-sample read depth of 1000 × for low-frequency (5%) variant detection. Point heteroplasmies below 10% frequency were confirmed through replicate amplification, and length heteroplasmy was quantitatively assessed using a custom read count analysis tool. Data analysis involved a redundant, dual-analyst review to minimize errors in haplotype reporting with additional QC checks performed by EMPOP. Applying these methods, eight sample sets were processed from five U.S. metapopulations (African American, Caucasian, Hispanic, Asian American, and Native American) corresponding to self-reported identity at the time of sample collection. Population analyses (e.g., haplotype frequencies, random match probabilities, and genetic distance estimates) were performed to evaluate the eight datasets, with over 95% of haplotypes unique per dataset. The platinum-quality mitogenome haplotypes presented in this study will enable forensic statistical calculations and thereby support the usage of mitogenome sequencing in forensic laboratories.

The impact of common PCR inhibitors on forensic MPS analysis.

Massively parallel sequencing holds great promise for new possibilities in the field of forensic genetics, enabling simultaneous analysis of multiple markers as well as offering enhanced short tandem repeat allele resolution. A challenge in forensic DNA analysis is that the samples often contain low amounts of DNA in a background that may interfere with downstream analysis. PCR inhibition mechanisms of some relevant molecules have been studied applying e.g. real-time PCR and digital PCR. However, a detailed understanding of the effects of inhibitory molecules on forensic MPS, including mechanisms and ways to relieve inhibition, is missing. In this study, the effects of two well-characterized PCR inhibitors, humic acid and hematin, have been studied using the ForenSeq DNA Signature Prep kit. Humic acid and hematin resulted in lowered read numbers as well as specific negative effects on certain markers. Quality control of libraries with Fragment analyzer showed that increasing amounts of inhibitors caused a lowered amplicon quantity and that the larger amplicons were more likely to drop out. Further, the inhibitor tolerance could be improved 5-10 times by addition of bovine serum albumin in the initial PCR. On the contrary to the samples with inhibitors, low-template samples resulted in lowered read numbers for all markers. This difference strengthened the conclusion that the inhibitors have a negative effect on the DNA polymerase activity in the initial PCR. Additionally, a common capillary gel electrophoresis-based STR kit was shown to handle at least 200 times more inhibitors than the ForenSeq DNA Signature Prep kit. This suggests that there is room for improvement of the PCR components to ensure analytical success for challenging samples, which is needed for a broad application of MPS for forensic STR analysis.

Sequence-based U.S. population data for 27 autosomal STR loci.

This manuscript reports Short Tandem Repeat (STR) sequence-based allele frequencies for 1036 samples across 27 autosomal STR loci: D1S1656, TPOX, D2S441, D2S1338, D3S1358, D4S2408, FGA, D5S818, CSF1PO, D6S1043, D7S820, D8S1179, D9S1122, D10S1248, TH01, vWA, D12S391, D13S317, Penta E, D16S539, D17S1301, D18S51, D19S433, D20S482, D21S11, Penta D, and D22S1045. Sequence data were analyzed by two bioinformatic pipelines and all samples have been evaluated for concordance with alleles derived from CE-based analysis at all loci. Each reported sequence includes high-quality flanking sequence and is properly formatted according to the most recent guidance of the International Society for Forensic Genetics. In addition, GenBank accession numbers are reported for each sequence, and associated records are available in the STRSeq BioProject (https://www.ncbi.nlm.nih.gov/bioproject/380127). The D3S1358 locus demonstrates the greatest average increase in heterozygosity across populations (approximately 10 percentage points). Loci demonstrating average increase in heterozygosity from 10 to 5 percentage points include (in descending order) D9S1122, D13S317, D8S1179, D21S11, D5S818, D12S391, and D2S441. The remaining 19 loci each demonstrate less than 5 percentage point increase in average heterozygosity. Discussion includes the utility of this data in understanding traditional CE results, such as informing stutter models and understanding migration challenges, and considerations for population sampling strategies in light of the marked increase in rare alleles for several of the sequence-based STR loci. This NIST 1036 data set is expected to support the implementation of STR sequencing forensic casework by providing high-confidence sequence-based allele frequencies for the same sample set which are already the basis for population statistics in many U.S. forensic laboratories.

Sequence-based US population data for the SE33 locus.

A set of 1036 U.S. Population Samples were sequenced using the Illumina ForenSeq DNA Signature Prep Kit. This sample set has been highly characterized using a variety of marker systems for human identification. The FASTQ files obtained from a ForenSeq DNA Signature Prep Kit experiment include several STR loci that are not reported in the associated software. These include SE33, DXS8377, DXS10148, DYS456, and DYS461. The sequence variation within the autosomal STR marker SE33 was evaluated using a customized bioinformatic approach to identify and characterize the locus in the 1036 data set. The analysis identified 53 unique alleles by length and 264 by sequence. An additional 10 alleles were detected when selected extended flanking regions were examined to resolve discordances. Allele frequencies and SE33 sequence motif patterns are reported for the 1036 data set. The comparison of numerical allele calls derived from sequence data to the allele calls obtained from commercial capillary electrophoresis-based STR typing kits resulted in 100% concordance, after manual data review and confirmation sequencing of three flanking region deletions. The analysis of this data set involved significant manual sequence curation and information support from length-based genotypes to ensure high confidence in the sequence-based allele calls. The challenges of interpreting the sequence data for SE33 consisted of high sequence noise, allele-size dependent variance in coverage, and heterozygote imbalance. As allele length increased, sequence depth of coverage and quality decreased at the terminal end. Accordingly, heterozygous genotype imbalance increased in proportion to increased distance between alleles.

Characterization of NIST human mitochondrial DNA SRM-2392 and SRM-2392-I standard reference materials by next generation sequencing.

Standard Reference Materials SRM 2392 and 2392-I are intended to provide quality control when amplifying and sequencing human mitochondrial genome sequences. The National Institute of Standards and Technology (NIST) offers these SRMs to laboratories performing DNA-based forensic human identification, molecular diagnosis of mitochondrial diseases, mutation detection, evolutionary anthropology, and genetic genealogy. The entire mtGenome (∼16569bp) of SRM 2392 and 2392-I have previously been characterized at NIST by Sanger sequencing. Herein, we used the sensitivity, specificity, and accuracy offered by next generation sequencing (NGS) to: (1) re-sequence the certified values of the SRM 2392 and 2392-I; (2) confirm Sanger data with a high coverage new sequencing technology; (3) detect lower level heteroplasmies (<20%); and thus (4) support mitochondrial sequencing communities in the adoption of NGS methods. To obtain a consensus sequence for the SRMs as well as identify and control any bias, sequencing was performed using two NGS platforms and data was analyzed using different bioinformatics pipelines. Our results confirm five low level heteroplasmy sites that were not previously observed with Sanger sequencing: three sites in the GM09947A template in SRM 2392 and two sites in the HL-60 template in SRM 2392-I.

Sequence variation of 22 autosomal STR loci detected by next generation sequencing.

Sequencing short tandem repeat (STR) loci allows for determination of repeat motif variations within the STR (or entire PCR amplicon) which cannot be ascertained by size-based PCR fragment analysis. Sanger sequencing has been used in research laboratories to further characterize STR loci, but is impractical for routine forensic use due to the laborious nature of the procedure in general and additional steps required to separate heterozygous alleles. Recent advances in library preparation methods enable high-throughput next generation sequencing (NGS) and technological improvements in sequencing chemistries now offer sufficient read lengths to encompass STR alleles. Herein, we present sequencing results from 183 DNA samples, including African American, Caucasian, and Hispanic individuals, at 22 autosomal forensic STR loci using an assay designed for NGS. The resulting dataset has been used to perform population genetic analyses of allelic diversity by length compared to sequence, and exemplifies which loci are likely to achieve the greatest gains in discrimination via sequencing. Within this data set, six loci demonstrate greater than double the number of alleles obtained by sequence compared to the number of alleles obtained by length: D12S391, D2S1338, D21S11, D8S1179, vWA, and D3S1358. As expected, repeat region sequences which had not previously been reported in forensic literature were identified.

Performance of a next generation sequencing SNP assay on degraded DNA.

Forensic DNA casework samples are often of insufficient quantity or quality to generate full profiles by conventional DNA typing methods. Polymerase chain reaction (PCR) amplification of short tandem repeat (STR) loci is inherently limited in samples containing degraded DNA, as the cumulative size of repeat regions, primer binding regions, and flanking sequence is necessarily larger than the PCR template. Additionally, traditional capillary electrophoresis (CE) assay design further inherently limits shortening amplicons because the markers must be separated by size. Non-traditional markers, such as single nucleotide polymorphisms (SNPs) and insertion deletion polymorphisms (InDels), may yield more information from challenging samples due to their smaller amplicon size. In this study, the performance of a next generation sequencing (NGS) SNP assay and CE-based STR, mini-STR, and InDel assays was evaluated with a series of fragmented, size-selected samples. Information obtained from the NGS SNP assay exhibited higher overall inverse random match probability (1/RMP) values compared to the CE-based typing assays, with particular benefit for fragment sizes ≤ 150 base pairs (bp). The InDel, mini-STR, and NGS SNP assays all had similar percentages of loci with reportable alleles at this level of degradation; however, the relatively fewer number of loci in the InDel and mini-STR assays results in the NGS SNP assay having at least nine orders of magnitude higher 1/RMP values. In addition, the NGS SNP assay and three CE-based assays (two STR and one InDel assay) were tested using a dilution series consisting of 0.5 ng, 0.1 ng, and 0.05 ng non-degraded DNA. All tested assays showed similar percentages of loci with reportable alleles at these levels of input DNA; however, due to the larger number of loci, the NGS SNP assay and the larger of the two tested CE-based STR assays both resulted in considerably higher 1/RMP values than the other assays. These results indicate the potential advantage of NGS SNP assays for forensic analysis of degraded DNA samples.

Collaborative EDNAP exercise on the IrisPlex system for DNA-based prediction of human eye colour.

The IrisPlex system is a DNA-based test system for the prediction of human eye colour from biological samples and consists of a single forensically validated multiplex genotyping assay together with a statistical prediction model that is based on genotypes and phenotypes from thousands of individuals. IrisPlex predicts blue and brown human eye colour with, on average, >94% precision accuracy using six of the currently most eye colour informative single nucleotide polymorphisms (HERC2 rs12913832, OCA2 rs1800407, SLC24A4 rs12896399, SLC45A2 (MATP) rs16891982, TYR rs1393350, and IRF4 rs12203592) according to a previous study, while the accuracy in predicting non-blue and non-brown eye colours is considerably lower. In an effort to vigorously assess the IrisPlex system at the international level, testing was performed by 21 laboratories in the context of a collaborative exercise divided into three tasks and organised by the European DNA Profiling (EDNAP) Group of the International Society of Forensic Genetics (ISFG). Task 1 involved the assessment of 10 blood and saliva samples provided on FTA cards by the organising laboratory together with eye colour phenotypes; 99.4% of the genotypes were correctly reported and 99% of the eye colour phenotypes were correctly predicted. Task 2 involved the assessment of 5 DNA samples extracted by the host laboratory from simulated casework samples, artificially degraded, and provided to the participants in varying DNA concentrations. For this task, 98.7% of the genotypes were correctly determined and 96.2% of eye colour phenotypes were correctly inferred. For Tasks 1 and 2 together, 99.2% (1875) of the 1890 genotypes were correctly generated and of the 15 (0.8%) incorrect genotype calls, only 2 (0.1%) resulted in incorrect eye colour phenotypes. The voluntary Task 3 involved participants choosing their own test subjects for IrisPlex genotyping and eye colour phenotype inference, while eye photographs were provided to the organising laboratory and judged; 96% of the eye colour phenotypes were inferred correctly across 100 samples and 19 laboratories. The high success rates in genotyping and eye colour phenotyping clearly demonstrate the reproducibility and the robustness of the IrisPlex assay as well as the accuracy of the IrisPlex model to predict blue and brown eye colour from DNA. Additionally, this study demonstrates the ease with which the IrisPlex system is implementable and applicable across forensic laboratories around the world with varying pre-existing experiences.

Comparison of base composition analysis and Sanger sequencing of mitochondrial DNA for four U.S. population groups.

A set of 711 samples from four U.S. population groups was analyzed using a novel mass spectrometry based method for mitochondrial DNA (mtDNA) base composition profiling. Comparison of the mass spectrometry results with Sanger sequencing derived data yielded a concordance rate of 99.97%. Length heteroplasmy was identified in 46% of samples and point heteroplasmy was observed in 6.6% of samples in the combined mass spectral and Sanger data set. Using discrimination capacity as a metric, Sanger sequencing of the full control region had the highest discriminatory power, followed by the mass spectrometry base composition method, which was more discriminating than Sanger sequencing of just the hypervariable regions. This trend is in agreement with the number of nucleotides covered by each of the three assays.

Allele frequencies for 40 autosomal SNP loci typed for US population samples using electrospray ionization mass spectrometry.

AIM: To type a set of 194 US African American, Caucasian, and Hispanic samples (self-declared ancestry) for 40 autosomal single nucleotide polymorphism (SNP) markers intended for human identification purposes. METHODS: Genotyping was performed on an automated commercial electrospray ionization time-of-flight mass spectrometer, the PLEX-ID. The 40 SNP markers were amplified in eight unique 5plex PCRs, desalted, and resolved based on amplicon mass. For each of the three US sample groups statistical analyses were performed on the resulting genotypes. RESULTS: The assay was found to be robust and capable of genotyping the 40 SNP markers consuming approximately 4 nanograms of template per sample. The combined random match probabilities for the 40 SNP assay ranged from 10-16 to 10-21. CONCLUSION: The multiplex PLEX-ID SNP-40 assay is the first fully automated genotyping method capable of typing a panel of 40 forensically relevant autosomal SNP markers on a mass spectrometry platform. The data produced provided the first allele frequencies estimates for these 40 SNPs in a National Institute of Standards and Technology US population sample set. No population bias was detected although one locus deviated from its expected level of heterozygosity.