If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Forensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Department of Human Biological Traces, Netherlands Forensic Institute, P.O. Box 24044, 2490 AA The Hague, The NetherlandsForensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Forensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Forensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Forensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Forensic Laboratory for DNA Research, Department of Human Genetics, Leiden University Medical Centre, Postzone S-05-P, P.O. Box 9600, 2300 RC Leiden, The Netherlands
Regularly, STR results obtained with different PCR amplification kits are compared, for instance with old cases, when revisiting cold cases or when addressing cross-border crimes. It is known that differences in primer design for the same loci in different kits may give rise to null alleles or shifted alleles. In this study, the genotyping results of 2085 Dutch male samples were compared for six autosomal STR kits (Promega's PowerPlex® 16, ESX-16 and ESI-17 Systems, Qiagen's Investigator® ESSplex Kit and Applied Biosystems’ AmpFlSTR® Identifiler and NGM PCR Amplification Kits). A total of 19 discordant autosomal genotyping results were obtained that were examined by sequence analysis using Roche-454 next generation sequencing and/or Sanger sequencing. A further 25 discordances were found and sequenced for the Amelogenin locus. The 24 samples showing the same primer binding site mutation at the Amelogenin locus were subjected to X-STR analysis in order to assess whether they could share a common origin, which appeared not to be the case. Based on the sequencing results, we set the final genotypes and determined the allele frequencies of 23 autosomal STRs for the Dutch reference database.
]. Examples of these kits are the PowerPlex ESX and ESI Systems (Promega Corporation (Promega), Madison, WI, USA), the Investigator ESSplex (Plus) Kit (Qiagen Benelux B.V. (Qiagen), Venlo, The Netherlands) and the AmpFlSTR NGM™ (SElect) PCR Amplification Kit (Applied Biosystems (AB), Foster City, CA, USA). When comparing genotyping results of the same donor obtained with different PCR amplification kits, differences such as null (a.k.a. silent) alleles and shifted alleles have been observed, primarily due to differences in primer design (e.g. [
European Network of Forensic Science Institutes (ENFSI): evaluation of new commercial STR multiplexes that include the European Standard Set (ESS) of markers.
]). With an increasing number of forensically available kits and a growing number of (international) DNA profile comparisons worldwide, for instance under the European Prüm Treaty, it is informative to know the extent of discordances at specific loci. When certain discordances occur regularly, this information may stimulate companies to include primer adjustments, such as degenerated primers, in newly developed forensic kits.
In this study, 2085 Dutch male samples were typed for six autosomal STR kits: the PowerPlex® 16 (PP16), ESX-16 and ESI-17 Systems (Promega), the Investigator® ESSplex Kit (ESS, Qiagen) and the AmpFlSTR® Identifiler and NGM PCR Amplification Kits (AB). We evaluated the concordancy of the genotyping results obtained with these kits. Discordant allele calls were examined using Roche-454 next generation sequencing (NGS) and/or Sanger sequencing in order to identify the causal mutations. Additional analyses with X-STRs were performed to analyse whether a set of 24 donors having the same Amelogenin primer binding site mutation could share a common origin. Based on the sequencing results, the final genotypes were set and used to create a new Dutch allele frequency database.
2. Material and methods
2.1 DNA samples, extraction and quantification
A total of 2085 male blood donors with self-declared Dutch ancestry were sampled from 99 locations across The Netherlands, while excluding major cities to avoid very recent admixture effects. All volunteers had given their informed consent, and a detailed description of the samples is given in [
Clinal distribution of human genomic diversity across the Netherlands despite archaeological evidence for genetic discontinuities in Dutch population history.
]. After anonymising the samples, DNA was robotically extracted either by the Autopure LS® system using the Gentra Puregene Blood Kit (Gentra Systems, Minneapolis, MN) or by the QIAcube using the QIAamp DNA Mini Kit (Qiagen Benelux B.V. (Qiagen), Venlo, The Netherlands). The samples were quantified with the Quantifiler® Duo DNA Quantification Kit on a 7500 Real-Time PCR System (Applied Biosystems (AB), Foster City, CA, USA).
2.2 PCR, capillary electrophoresis and DNA profile analysis
DNA amplifications were performed using six different autosomal STR kits: PP16, ESX-16 and ESI-17 (Promega), ESS (Qiagen) and Identifiler and NGM (AB). PCR amplification programs were performed according to the manufacturer's protocols, except for the reaction volumes (Table 1). The PCR input and reaction volume, the capillary electrophoresis instrument, setup and settings, and the DNA profile analysis software and detection threshold are shown in Table 1 for the six different STR kits.
Table 1Characteristics of the PCR amplification, capillary electrophoresis and DNA profile analysis for the six autosomal PCR kits analysed in this study.
CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Clinal distribution of human genomic diversity across the Netherlands despite archaeological evidence for genetic discontinuities in Dutch population history.
], and no substructure was detected (data not shown). Therefore, the complete dataset was interpreted as one group. The 2085 genotypes from all six kits were compared using Excel and discordant allele calls were identified. After sequencing analysis of discordant alleles, one final database was created comprising 2085 genotypes for 23 autosomal STRs and Amelogenin. With this database, the allele frequencies and descriptive statistics for this Dutch population sample were determined using the Excel Microsatellite Toolkit [
The 2085 DNA samples were analysed with six autosomal STR kits (PP16, ESX-16, ESI-17, ESS, Identifiler and NGM), comprising a total of 23 different autosomal STR loci and Amelogenin (Fig. 1). The genotyping results for the 20 STRs and Amelogenin that are present in more than one kit (Penta D and E only reside in PP16, SE33 only resides in ESI-17; see Fig. 1) were compared and 44 discordances were detected in 38 DNA samples (Table 2). These discrepancies comprise 43 null alleles on 12 loci and one shifted allele with a size difference of one nucleotide (Table 2). Just when all 2085 samples had been analysed with the NGM kit, Applied Biosystems adjusted the kit (without changing the name; kit lot-numbers 1103010 M and up) by adding extra primers for Amelogenin, D2S441 and D22S1045 that cover reported primer binding site mutations on these loci [
]. All discordant samples were reanalysed with this new version of NGM (here denoted NGM2), and the 24 missing X alleles on Amelogenin and the absent allele 14.1 on D2S441 were now detected and concordant with the results from other kits. As a result, the final number of discordant results decreased to 19, and these discordances occurred in 13 of the 2085 samples (for six samples, the same discordant result was seen in two different kits; Table 2). Seven discordant alleles were found for Identifiler and NGM/NGM2, three for ESI-17, and one for both PP16 and ESS (Table 2); ESX-16 showed no discordant allele calls (Table 2).
Fig. 1Fragment lengths for 23 autosomal STR markers and Amelogenin and their presence in six forensic STR kits: PP16, ESX-16 and ESI-17 (Promega, grey shades), ESS (Qiagen, red shade), Identifiler and NGM (Applied Biosystems, blue shades). Amplicon sizes are based on the shortest and longest fragment length in the allelic ladder for each marker. When a kit name is presented in grey, the marker is not present in the kit.
Table 2Discordancy between six autosomal STR kits: PP16, ESX-16, ESI-17, ESS, Identifiler and NGM, based on 2085 Dutch male donor samples. Discordant results are shown in grey.
Both observations involve the same donor and the discordancy was seen in two different kits.
454
B: 52 A>T
19 nt after start Identifiler + NGM F-primer
Total
1
0
3
1
7
32
7
44
19
a NGM2 is only tested on the discordant samples.
b Both observations involve the same donor and the discordancy was seen in two different kits.
c Mutation position before (B) or after (A) the repeat, and in the direction as given in STRbase (accessed October 2013); see also Supplementary Data 1.
d For most primers, the (start) position is given in Supplementary Data 1.
3.2 Roche-454 NGS and/or Sanger sequencing of samples with discordant results
In order to determine the genetic variations causing the above-mentioned 44 discordant results, the loci involved were sequenced for the affected samples and control samples. First, Roche-454 NGS was applied using primers designed by Kline et al. [
], and when no (clear) results were obtained, Sanger sequencing was used with custom-designed primers both further away from the repeat than those by Kline et al. [
] (Supplementary Table 1). For 42 of the 44 discordances a nucleotide change or nucleotide insertion or deletion was observed that probably caused the discordant genotyping result (Table 2 and Supplementary Data 1). For two samples, with a null allele at D12S391 for the ESI-17 and NGM/NGM2 kits, the discordant allele could not be amplified and no sequence change was revealed. There is a clear relation, however, between the occurrence of the null allele and the location of the various primers: kits that do detect the allele have reverse primers close to the repeat structure (for ESX the reverse primer starts 30 nt and for ESS 43 nt after the repeat; Supplementary Data 1), while the kits and custom-designed primers that do not reveal the allele have reverse primers further on (primer start for NGM is at position 125 (Supplementary Data 1) and for the various sequencing primers at nucleotide 213 [
], 271 or 345). We infer that both null alleles are caused by a deletion of at least a few hundred nucleotides, residing after the position of the ESS reverse primer and covering the binding sites of the sequencing primers and the reverse primers in NGM/NGM2 and ESI-17 (for this kit, the exact primer start is unknown).
Many studies have been conducted to determine the molecular basis of discordant results (e.g. [
Detection of a primer-binding site polymorphism for the STR locus D16S539 using the Powerplex 1.1 system and validation of a degenerate primer to correct for the polymorphism.
]) and are due to a primer binding site mutation, manufacturers may add degenerate primers to the commercially available STR kits in order to recover the specific allele [
Detection of a primer-binding site polymorphism for the STR locus D16S539 using the Powerplex 1.1 system and validation of a degenerate primer to correct for the polymorphism.
]. The majority of discordant results, however, are rare, like most of the discordances found in our study. Despite their rarity, at least three of the autosomal discordances that we detected have been described before: 1) a G>A substitution 172 bp after the D18S51 repeat motif, described as position 75615 in GenBank sequence AC021803 by Delamoye et al. [
Identification of new primer binding site mutations at TH01 and D13S317 loci and determination of their corresponding STR alleles by allele-specific PCR.
]. Notwithstanding, null alleles were detected for D2S1338 and D16S539 for Identifiler that were not observed for NGM (Table 2). We think the prolonged annealing time in the NGM protocol (3 min versus 1 min) made it possible to detect these alleles, albeit with relatively low peak heights compared to the other allele in the heterozygous allele set (peak height ratios were 0.06 and 0.13, respectively).
All 24 samples that showed a null allele for X on Amelogenin when typed with NGM have the same G>A substitution (Table 2 and Supplementary Data 1). Green et al. [
] described that the discovery of rare population-specific variant alleles for AMEL prompted the adjustment of the AMEL primers, which was put into effect in the summer of 2011. The use of this updated NGM kit (here denoted NGM2), resolved all AMELX null alleles in our samples, indicating that this mutation is not very rare in the Dutch dataset as it occurs with a frequency of 1.15%. Interestingly, none of the other reported AMELX null alleles [
], with estimated frequencies up to 2% in specific populations, were found in our dataset. In order to analyse whether the Dutch male donors of these samples could share a common origin, we tested X-STRs in all these Amelogenin mutation carriers. A homemade multiplex was used that contains the SRY male gender marker and 13 X-STRs of all four linkage groups [
]. Three of the markers in this multiplex (DXS8378, DXS9902 and DXS6807) reside in the same area of the X-chromosome as Amelogenin, but no apparent similarities at these three loci were seen among the 24 samples (results not shown). Thus, no apparent association between the X-STR results and the G>A substitution in the Amelogenin gene could be made. This suggests there is no recent common origin for this mutation in our study population.
3.3 Number and percentage discordant and concordant results
A consolidated database (denoted 2085-database) comprising 2085 genotypes of 23 autosomal STR loci and Amelogenin was created. For the samples that showed a null allele, the detected allele was included in the database. For the sample with the shifted allele in ESS for D19S433, we included the allele call presented by the other four kits, as the deleted nucleotide was found in the flanking regions of the repeat (Table 2 and Supplementary Data 1). Next, we determined the number and percentage of discordant genotypes and alleles for each locus (Table 3). The highest percentage of discordant results was found for locus D12S391 (genotypes: 0.048% and alleles: 0.024%), which was typed by four different kits. Fully concordant results for all kits tested were obtained for 9 loci typed with 2 to 6 kits. For three loci (Penta D, Penta E and SE33) no comparisons could be made, as they were only present in one of the tested kits. Overall, 99.991% of the genotypes and 99.995% of the alleles show fully concordant results for our Dutch reference dataset of 2085 samples.
Table 3Number and percentage of discordant results on genotype and allele level obtained with six autosomal STR kits.
], which was based on 231 samples for the SGM Plus (AB) and 201 for the Profiler (AB) loci, 52 new alleles were detected that had between one and thirteen occurrences in the 2085-dataset. One allele that was detected once in the former database was not revealed in the 2085-dataset. The largest difference between the allele frequencies of the former and the new database is 0.048. In the new allele frequency database 97 of the 361 different alleles have frequencies below 0.001 (≤ four occurrences) of which 48 alleles occur only once.
Table 4Overview of the number and frequency of alleles per marker.
The summary statistics together with the heterozygosity and PIC values are shown in Supplementary Table 2. SE33 shows the highest heterozygosity and PIC values, which can be explained by the high number of microvariants resulting in 59 different alleles (Table 4) detected for this locus compared to the average of 15.65 different alleles per marker.
4. Concluding remarks
Nineteen discordant alleles were detected in 2085 Dutch male donor samples that were analysed for six autosomal STR kits (giving 99.995% concordant alleles overall). The numbers of null or shifted alleles vary for the different commercial kits: seven in Identifiler, seven in NGM2, three in ESI-17, one in PP16, one in ESS and none in ESX-16. Upon Roche-454 and/or Sanger sequencing nucleotide changes, insertions and deletions (single nucleotide or stretches of at least few hundred nucleotides) were found that could account for the discordant genotypes. The use of as much as 2085 samples generated a highly detailed allele frequency database, which can be of assistance in evidentiary value calculations.
Acknowledgements
This study was supported by a grant from The Netherlands Genomics Initiative/Netherlands Organization for Scientific Research (NWO) within the framework of the Forensic Genomics Consortium Netherlands. The authors would like to thank Applied Biosystems, Promega and Qiagen for their unconditional in kind support, and Arnoud Kal for critically reading the manuscript.
European Network of Forensic Science Institutes (ENFSI): evaluation of new commercial STR multiplexes that include the European Standard Set (ESS) of markers.
Clinal distribution of human genomic diversity across the Netherlands despite archaeological evidence for genetic discontinuities in Dutch population history.
CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Detection of a primer-binding site polymorphism for the STR locus D16S539 using the Powerplex 1.1 system and validation of a degenerate primer to correct for the polymorphism.
Identification of new primer binding site mutations at TH01 and D13S317 loci and determination of their corresponding STR alleles by allele-specific PCR.
☆This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.
00024-6&id=gr1.jpg){kind=link}