Claus Børsting

Next generation sequencing and its applications in forensic genetics

It has been almost a decade since the first next generation sequencing (NGS) technologies emerged and quickly changed the way genetic research is conducted. Today, full genomes are mapped and published almost weekly and with ever increasing speed and decreasing costs. NGS methods and platforms have matured during the last 10 years, and the quality of the sequences has reached a level where NGS is used in clinical diagnostics of humans. Forensic genetic laboratories have also explored NGS technologies and especially in the last year, there has been a small explosion in the number of scientific articles and presentations at conferences with forensic aspects of NGS. These contributions have demonstrated that NGS offers new possibilities for forensic genetic case work. More information may be obtained from unique samples in a single experiment by analyzing combinations of markers (STRs, SNPs, insertion/deletions, mRNA) that cannot be analyzed simultaneously with the standard PCR-CE methods used today. The true variation in core forensic STR loci has been uncovered, and previously unknown STR alleles have been discovered. The detailed sequence information may aid mixture interpretation and will increase the statistical weight of the evidence. In this review, we will give an introduction to NGS and single-molecule sequencing, and we will discuss the possible applications of NGS in forensic genetics.

Multiplex PCR and minisequencing of SNPs— a model with 35 Y chromosome SNPs

We have developed a robust single nucleotide polymorphism (SNPs) typing assay with co-amplification of 25 DNA-fragments and the detection of 35 human Y chromosome SNPs. The sizes of the PCR products ranged from 79 to 186 base pairs. PCR primers were designed to have a theoretical Tm of 60 +/- 5 degrees C at a salt concentration of 180 mM. The sizes of the primers ranged from 19 to 34 nucleotides. The concentration of amplification primers was adjusted to obtain balanced amounts of PCR products in 8mM MgCl2. For routine purposes, 1 ng of genomic DNA was amplified and the lower limit was approximately 100 pg DNA. The minisequencing reactions were performed simultaneously for all 35 SNPs with fluorescently labelled dideoxynucleotides. The size of the minisequencing primers ranged from 19 to 106 nucleotides. The minisequencing reactions were analysed by capillary electrophoresis and multicolour fluorescence detection. Female DNA did not influence the results of Y chromosome SNP typing when added in concentrations more than 300 times the concentrations of male DNA. The frequencies of the 35 SNPs were determined in 194 male Danes. The gene diversity of the SNPs ranged from 0.01 to 0.5.

High-throughput sequencing of core STR loci for forensic genetic investigations using the Roche Genome Sequencer FLX platform.

The analysis and profiling of short tandem repeat (STR) loci is routinely used in forensic genetics. Current methods to investigate STR loci, including PCR-based standard fragment analyses and capillary electrophoresis, only provide amplicon lengths that are used to estimate the number of STR repeat units. These methods do not allow for the full resolution of STR base composition that sequencing approaches could provide. Here we present an STR profiling method based on the use of the Roche Genome Sequencer (GS) FLX to simultaneously sequence multiple core STR loci. Using this method in combination with a bioinformatic tool designed specifically to analyze sequence lengths and frequencies, we found that GS FLX STR sequence data are comparable to conventional capillary electrophoresis-based STR typing. Furthermore, we found DNA base substitutions and repeat sequence variations that would not have been identified using conventional STR typing.

Second-generation sequencing of forensic STRs using the Ion Torrent™ HID STR 10-plex and the Ion PGM™

Second-generation sequencing (SGS) using Roche/454 and Illumina platforms has proved capable of sequencing the majority of the key forensic genetic STR systems. Given that Roche has announced that the 454 platforms will no longer be supported from 2015, focus should now be shifted to competing SGS platforms, such as the MiSeq (Illumina) and the Ion Personal Genome Machine (Ion PGM™; Thermo Fisher). There are currently several challenges faced with amplicon-based SGS STR typing in forensic genetics, including current lengths of amplicons for CE-typing and lack of uniform data analysis between laboratories. Thermo Fisher has designed a human identification (HID) short tandem repeat (STR) 10-plex panel including amelogenin, CSF1PO, D16S539, D3S1358, D5S818, D7S820, D8S1179, TH01, TPOX and vWA, where the primers have been designed specifically for the purpose of SGS and the data analysis is supported by Ion Torrent™ software. Hence, the combination of the STR 10-plex and the Ion PGM™ represents the first fully integrated SGS STR typing solution from PCR to data analysis. In this study, four experiments were performed to evaluate the alpha-version of the STR 10-plex: (1) typing of control samples; (2) analysis of sensitivity; (3) typing of mixtures; and (4) typing of biological crime case samples. Full profiles and concordant results between replicate SGS runs and CE-typing were observed for all control samples. Full profiles were seen with DNA input down to 50 pg, with the exception of a single locus drop-out in one of the 100 pg dilutions. Mixtures were easily deconvoluted down to 20:1, although alleles from the minor contributor had to be identified manually as some signals were not called by the Ion Torrent™ software. Interestingly, full profiles were obtained for all biological samples from real crime and identification cases, in which only partial profiles were obtained with PCR-CE assays. In conclusion, the Ion Torrent™ HID STR 10-plex panel offers an all-in-one solution from amplification of STRs and amelogenin, and sequencing to data analysis.

Performance of the SNPforID 52 SNP-plex assay in paternity testing

The performance of a multiplex assay with 52 autosomal single nucleotide polymorphisms (SNPs) developed for human identification was tested on 124 mother-child-father trios. The typical paternity indices (PIs) were 10(5)-10(6) for the trios and 10(3)-10(4) for the child-father duos. Using the SNP profiles from the randomly selected trios and 700 previously typed individuals, a total of 83,096 comparisons between mother, child and an unrelated man were performed. On average, 9-10 mismatches per comparison were detected. Four mismatches were genetic inconsistencies and 5-6 mismatches were opposite homozygosities. In only two of the 83,096 comparisons did an unrelated man match perfectly to a mother-child duo, and in both cases the PI of the true father was much higher than the PI of the unrelated man. The trios were also typed for 15 short tandem repeats (STRs) and seven variable number of tandem repeats (VNTRs). The typical PIs based on 15 STRs or seven VNTRs were 5-50 times higher than the typical PIs based on 52 SNPs. Six mutations in tandem repeats were detected among the randomly selected trios. In contrast, there was not found any mutations in the SNP loci. The results showed that the 52 SNP-plex assay is a very useful alternative to currently used methods in relationship testing. The usefulness of SNP markers with low mutation rates in paternity and immigration casework is discussed.

Forensic validation of the SNPforID 52-plex assay

The advantages of single nucleotide polymorphism (SNP) typing in forensic genetics are well known and include a wider choice of high-throughput typing platforms, lower mutation rates, and improved analysis of degraded samples. However, if SNPs are to become a realistic supplement to current short tandem repeat (STR) typing methods, they must be shown to successfully and reliably analyse the challenging samples commonly encountered in casework situations. The European SNPforID consortium, supported by the EU GROWTH programme, has developed a multiplex of 52 SNPs for forensic analysis, with the amplification of all 52 loci in a single reaction followed by two single base extension (SBE) reactions which are detected with capillary electrophoresis. In order to validate this assay, a variety of DNA extracts were chosen to represent problems such as low copy number and degradation that are commonly seen in forensic casework. A total of 40 extracts were used in the study, each of which was sent to two of the five participating laboratories for typing in duplicate or triplicate. Laboratories were instructed to carry out their analyses as if they were dealing with normal casework samples. Results were reported back to the coordinating laboratory and compared with those obtained from traditional STR typing of the same extracts using Powerplex 16 (Promega). These results indicate that, although the ability to successfully type good quality, low copy number extracts is lower, the 52-plex SNP assay performed better than STR typing on degraded samples, and also on samples that were both degraded and of limited quantity, suggesting that SNP analysis can provide advantages over STR analysis in forensically relevant circumstances. However, there were also additional problems arising from contamination and primer quality issues and these are discussed.

Validation of a single nucleotide polymorphism (SNP) typing assay with 49 SNPs for forensic genetic testing in a laboratory accredited according to the ISO 17025 standard

A multiplex assay with 49 autosomal single nucleotide polymorphisms (SNPs) developed for human identification was validated for forensic genetic casework and accredited according to the ISO 17025 standard. The multiplex assay was based on the SNPforID 52plex SNP assay [J.J. Sanchez, C. Phillips, C. Børsting, K. Balogh, M. Bogus, M. Fondevila, C.D. Harrison, E. Musgrave-Brown, A. Salas, D. Syndercombe-Court, P.M. Schneider, A. Carracedo, N. Morling, A multiplex assay with 52 single nucleotide polymorphisms for human identification, Electrophoresis 27 (2006) 1713-1724], where 52 fragments were amplified in one PCR reaction. The SNPs were analysed by single base extension (SBE) and capillary electrophoresis. Twenty-three of the original SBE primers were altered to improve the overall robustness of the assay and to simplify the analysis of the SBE results. A total of 216 samples from 50 paternity cases and 33 twin cases were typed at least twice for the 49 SNPs. All electropherograms were analysed independently by two expert analysts prior to approval. Based on these results, detailed guidelines for analysis of the SBE products were developed. With these guidelines, the peak height ratio of a heterozygous allele call or the signal to noise ratio of a homozygous allele call is compared with previously obtained ratios. A laboratory protocol for analysis of SBE products was developed where allele calls with unusual ratios were highlighted to facilitate the analysis of difficult allele calls. The guidelines for allele calling proved to be highly efficient for the detection of DNA mixtures and contaminated DNA preparations. DNA from two individuals was mixed in seven different ratios ranging from 1:1 to 1:10; all mixtures were easily identified as mixtures.

Evaluation of the Ion Torrent™ HID SNP 169-plex: A SNP typing assay developed for human identification by second generation sequencing

The Ion Torrent™ HID SNP assay amplified 136 autosomal SNPs and 33 Y-chromosome markers in one PCR and the markers were subsequently typed using the Ion PGM™ second generation sequencing platform. A total of 51 of the autosomal SNPs were selected from the SNPforID panel that is routinely used in our ISO 17025 accredited laboratory. Concordance between the Ion Torrent™ HID SNP assay and the SNPforID assay was tested by typing 44 Iraqis twice with the Ion Torrent™ HID SNP assay. The same samples were previously typed with the SNPforID assay and the Y-chromosome haplogroups of the individuals were previously identified by typing 45 Y-chromosome SNPs. Full concordance between the assays were obtained except for the SNP genotypes of two SNPs. These SNPs were among the eight SNPs (rs2399332, rs1029047, rs10776839, rs4530059, rs8037429, rs430046, rs1031825 and rs1523537) with inconsistent allele balance among samples. These SNPs should be excluded from the panel. The optimal amount of DNA in the PCR seemed to be ≥0.5ng. Allele drop-outs were rare and only seen in experiments with <0.5ng input DNA and with a coverage of <50reads. No allele drop-in was observed. The great majority of the heterozygote allele balances were between 0.6 and 1.6, which is comparable to the heterozygote balances of STRs typed with PCR-CE. The number of reads with base calls that differed from the genotype call was typically less than five. This allowed detection of 1:100 mixtures with a high degree of certainty in experiments with a high total depth of coverage. In conclusion, the Ion PGM™ is a very promising platform for forensic genetics. However, the secondary sequence analysis software made wrong genotype calls from correctly sequenced alleles. These types of errors must be corrected before the platform can be used in case work. Furthermore, the sequence analysis software should be further developed and include quality settings for each SNP based on validation studies.

Forensic typing of autosomal SNPs with a 29 SNP-multiplex—Results of a collaborative EDNAP exercise

We report the results of an inter-laboratory exercise on typing of autosomal single nucleotide polymorphisms (SNP) for forensic genetic investigations in crime cases. The European DNA Profiling Group (EDNAP), a working group under the International Society for Forensic Genetics (ISFG), organised the exercise. A total of 11 European and one US forensic genetic laboratories tested a subset of a 52 SNP-multiplex PCR kit developed by the SNPforID consortium. The 52 SNP-multiplex kit amplifies 52 DNA fragments with 52 autosomal SNP loci in one multiplex PCR. The 52 SNPs are detected in two separate single base extension (SBE) multiplex reactions with 29 and 23 SNPs, respectively, using SNaPshot kit, capillary electrophoresis and multicolour fluorescence detection. For practical reasons, only the 29 SBE multiplex reaction was carried out by the participating laboratories. A total of 11 bloodstains on FTA cards including a sample of poor quality and a negative control were sent to the laboratories together with the essential reagents for the initial multiplex PCR and the multiplex SBE reaction. The total SNP locus dropout rate was 2.8% and more than 50% of the dropouts were observed with the poor quality sample. The overall rate of discrepant SNP allele assignments was 2.0%. Two laboratories reported 60% of all the discrepancies. Two laboratories reported all 29 SNP alleles in all 10 positive samples correctly. The results of the collaborative exercise were surprisingly good and demonstrate that SNP typing with SBE, capillary electrophoresis and multicolour detection methods can be developed for forensic genetics.

Characterization of mutations and sequence variants in the D21S11 locus by next generation sequencing.

We sequenced the D21S11 locus in 77 individuals from Danish paternity cases using 454 FLX next generation sequencing (NGS) technology. All samples were also typed with the AmpFlSTR Profiler Plus or the AmpFlSTR Identifiler PCR Amplification kits as part of paternity investigations. In 18 of the confirmed trios, a genetic inconsistency was observed between one of the parents and the child at the D21S11 locus. NGS of the D21S11 locus revealed which allele had mutated from which parent to the child in 13 of these trios. All characterized mutations could be explained by single-step mutations in the longest sub-repeat of D21S11. A total of 53 of the 77 sequenced samples originated from unrelated individuals. Twenty different D21S11 alleles were detected by NGS in these individuals whereas only 13 different alleles were observed with fragment analysis. Several alleles had the same lengths but different sequences, e.g. four and three different alleles were detected by NGS with lengths determined by CE corresponding to allele 30 and allele 31, respectively.