Introduction

Polymorphic short tandem repeats (STR) are powerful and widely used genetic markers for paternity testing. Some laboratories have only used STRs for this purpose. In forensic casework, ordinary parentage cases can be expected to be solved by a commercially available multiplexes of STRs genotyping kit, e.g., PowerPlex 16 System Kit (Promega, Madison, WI, USA) or AmpFℓSTR Identifiler polymerase chain reaction (PCR) amplification kit (Applied Biosystems, Foster City, CA, USA). However, for more difficult cases, such as deficient cases [1], related individuals are involved [2, 3], mutations are encountered [4], or complex kinship analysis [5, 6] and common STRs cannot provide enough discriminatory power [5, 7, 8]. Additional STRs are needed to complement conventional analysis for obtain more information. In this paper, we present the allele frequencies, forensic efficiency values of nine non-combined DNA index system (CODIS) autosomal STR loci which were analyzed in a Chinese population using a new multiplex PCR system.

Materials and methods

Five hundred six bloodstains samples deposited on qualitative filter papers were collected from unrelated Chinese Han individuals residing in Guangdong Province, Southern China. Genomic DNA was extracted from bloodstains using a Chelex-100 method. Of the bloodstain in 250 μl 5% Chelex-100 (w/v), 3 × 3-mm size cutting was incubated at 56°C for 2 h to overnight. Subsequently, the extract was heated at 98°C for 8 min, vortexed for 10 s, and centrifuged in a microcentrifuge.

Amplification of STRs was carried out using a multiplex PCR system STRtyper-10F/G kit (Condon, Zhuhai, China). This kit comprises nine STR loci and a sex determination gene: D2S1772, D6S1043, D7S3048, D8S1132, D11S2368, D12S391, D13S325, D18S1364, GATA198B05, and Amelogenin. The genomic mapping information and genotypes of DNA from cell line 9947A (Promega, Madison, WI, USA) are presented in Table 1. These loci and their alleles have been confirmed by cloning sequencing. Multiplex PCR of the STRtyper-10F/G kit was performed using 2 µl of template DNA in 10 µl volume including 1 µl 10× Primer Set, 2 µl 5× PCR Reaction Mix, and 0.2 µl (1 U) HS-Taq DNA polymerase. PCR cycle condition was 95°C for 5 min, followed by 28 cycles at 94°C for 1 min, 60°C for 1 min, 72°C for 1 min, and final extension at 72°C for 20 min. All amplifications were done on GeneAmp PCR 9700 (Applied Biosystems, Foster City, CA, USA) or MJ PTC-100 (Bio-Rad, Hercules, CA, USA) thermal cyclers.

Table 1 Physical and genetic mapping information of the studied nine short tandem repeat (STRs) and commonly used STRs on the same chromosome and genotyping of 9947 A DNA of nine STR markers in this study
Full size table

The PCR products were separated and detected by capillary electrophoresis using an ABI 3100 Genetic Analyzer (Applied Biosystems) according to the manufacturer's instructions. Briefly, 1 µl PCR product or Allelic Ladder was mixed with 10 µl Hi-Di formamide (with 5% internal lane standard), denatured at 95°C for 3 min, followed by chilling on ice for 3 min immediately. Then, the mixture was analyzed on the ABI 3100 Genetic Analyzer with Data Collection Software (version 2.0); POP™-4 (Applied Biosystems) was utilized for separations on a 36-cm array. Samples were injected electrokinetically for 10 s at 3 kV. Allele calls were made in GeneMapper ID v3.2 software by comparison with kit allelic ladders using the panel and bin files supplied by the kit manufacturer. If an off-ladder peak was encountered, the sample was typed repeatedly to confirm the typing. New allele was determined as described in [9] and added to the allelic bin set. Control DNA from cell line 9947A (Promega, Madison, WI, USA) was genotyped as standard reference in all experiments.

Allelic frequencies, expected heterozygosity, probability of exclusion in trios (PET), probability of exclusion in duos (PED), and probability of match (PM) were computed using GenAlEx (ver6.2) software package [10], Power of discrimination (PD) were calculated using formula PD = 1 − PM. The exact test (base on 5,000 shufflings) for Hardy–Weinberg equilibrium and linkage disequilibrium was carried out by using GDA ver. 1.1 (http://lewis.eeb.uconn.edu/lewishome/software.html) software. Exact test differentiation of allele frequency distribution between populations was performed by using Arlequin ver 3.11 software package [11].

Results and discussion

All samples showed completed genotyping profiles. The plot of Allelic Ladder and a sample profile was displayed in Fig. S1. Microvariant such as N.1, N.2, and N.3 alleles were detected at loci D6S1043, D7S3048, D8S1132, D11S2368, D12S391, and D18S1364. The full typing data including PowerPlex 16 System is presented in Table S1. Allele frequencies and forensic parameters are presented in Table 2. Genetic polymorphism of D2S1772 [12], D6S1043 [12], D7S3048 [12], D8S1132 [7, 13, 14], D12S391 [8, 13–15], and D13S325 [16] have been published previously. Comparison of allele frequency distribution between the studied population and other published population was performed. The results showed significant differentiation between our data and other populations except Chinese Mongolian at D2S1772, German at D8S1132, and Southeast China Han at D12S391 (Table S2). D11S2368, D18S1364, and GATA198B05 have not been reported for forensic purpose. Data in Table 2 reveals the three loci have highly polymorphic information.

Table 2 Allele frequencies and forensic parameters for the nine non-combined DNA index system short tandem repeats used in this study
Full size table

Base on the exact test, only the locus D11S2368 showed a departure from Hardy–Weinberg expectation (p = 0.0310), but it was well below the significance threshold after employing a Bonferroni's correction [17].

Pairwise linkage disequilibrium tests were performed for all possible two-locus combinations, and one deviation was showed in 36 pairs (D2S1772/GATA198B05, p = 0.0456), which is not to be significant after Bonferroni's correction. Considering the loci as independent, total PD for the nine loci is 0.999999999999753. Combined probability of exclusion in trios and combined probability of exclusion in duos is 0.999981 and 0.999025, respectively.

Besides the nine loci mentioned above, more loci may need in parentage analysis. So, samples were typed using PowerPlex 16 system simultaneously. Exact test indicated that only Penta D (p = 0.0216) and Penta E (p = 0.0417) did not meet Hardy–Weinberg expectations for the loci in PowerPlex 16 system. Pairwise linkage disequilibrium tests were also performed for all pairs of two loci, including the loci comprising in PowerPlex 16 System. Although some markers were located on the same chromosome (Table 1), linkage disequilibrium between these loci was not observed. Five deviations (including one described above) were detected in 276 pairwise comparisons (D2S1772/GATA198B05, p = 0.0456; D3S1358/GATA198B05, p = 0.0184; D11S2368/Penta D, p = 0.0366; D13S325/D3S1358, p = 0.0488; and D18S1364/Penta D, p = 0.0004), with p values below the nominal 0.05 threshold. However, none of them remained significant after the Bonferroni's correction for multiple testing (resulting in a significance threshold p value of 0.0002). Furthermore, these markers of pairwise of deviations do not locate on the same chromosome. Overall, there is no linkage disequilibrium between the nine loci studied here and the Powerplex 16 system loci.

It is important to note that in cases where a mutation might have occurred, another mutation may arise when more STRs were used in parentage analysis. Therefore, mutation rate of the nine non-CODIS STRs were investigated. In total of 51 mother/father/child trios and 432 parent/child duos, single mismatch was founded in seven cases. The observed mutation rates ranged from 0 to 5.618 × 10−3 (Table 3) and are in the range reported by Brinkmann et al. [18], but the sample size is too small to draw a reliable mutation rate.

Table 3 Observed mutations in a total of 51 mother/father/child trios and 432 parent/child duos for nine non-combined DNA index system short tandem repeats
Full size table

We used aliquots of the cell line 9947A DNA (200, 300, 400, 500, 750, 1,000, 1,500, 2,000 2,500, and 3,000 pg) to determine the minimum quantity of DNA required to achieve reliable results. With 500–1,500 pg of template DNA, allele typing at all nine STRs and Amel gene was very successful (peak highs at all loci exceeded 100 relative fluorescence units). Out of this range, unbalance or artifact peaks maybe appear.

In conclusion, our work has demonstrated the nine non-CODIS loci are highly polymorphic markers and highly discriminating tools. These markers are probably unlinked to common STRs included in commercially available kits (e.g., PowerPlex 16 system) and useful to obtain additional information in STR analysis for forensic casework, especially useful for complex kinship analysis, paternity testing in deficiency cases or within related individuals. In our practice, we have used the nine non-CODIS loci to solve the complicated cases, including all cases of parent/child duos and cases of suspected mutation.