Introduction

Male-specific Y chromosome short tandem repeats (Y-STRs) are particularly important genetic markers located on the Y chromosome. Males within the same paternal lineage typically share identical Y-STR haplotypes, except for mutations. Therefore, Y-STRs not only are valuable in studies of genetic lineages, human evolution, and population history1,2,3,4,5 but also have unique applications in forensic medicine, including the investigation of sexual offenses, paternal kinship identification, family pedigree tracing, and inference of ancestral geography6,7,8.

In 1992, Roewer9 reported the polymorphism of Y-STR, and in the same year, it was applied in forensic practice10. Since then, additional Y-STRs have been incorporated into forensic Y-STR haplotype analysis. Some commercial kits, such as the Yfiler, demonstrate high polymorphism and individual discrimination power across diverse populations, exhibiting excellent stability, sensitivity, and accuracy for forensic applications11. However, because of the paternal inheritance of the male-specific part of the Y chromosome and the relatively low mutation rate (10− 4‒10− 3) of the Y-STRs typically used in forensic applications, their ability to differentiate individuals may be limited to inbred populations, populations with bottleneck effects, or male relative differentiation12,13,14,15,16.

One of the critical tasks of forensic DNA analysis is individual identification. Y-STRs with a high mutation rate are expected to overcome or at least reduce the above limitations. In 2010, Ballantyne et al.17 identified 14 loci (DYS627, DYS576, DYF387S1, DYS518, DYS570, DYS449, DYF404S1, DYF399S1, DYS547, DYS526a/b, DYS626, DYF403S1a, DYF403S1b, DYS612) with high mutation rates (> 10− 2) and termed these rapidly mutating Y-STRs (RM Y-STRs). In studies involving 604 unrelated males from 51 populations worldwide18, as well as 2,378 father‒son pairs19, the haplotype discrimination capacity of Yfiler was 90.4%, with a father‒son differentiation rate of 4.5%. In contrast, the 14 RM Y-STRs presented a haplotype discrimination capacity of 98.3%, with a father–son differentiation rate of 26.9%. Subsequent studies in Turkey20, Pakistan21, and China22,23,24,25 have demonstrated that RM Y-STRs have significant value for differentiating both related and unrelated males. In 2020, Ralf et al.26 identified 12 novel RM Y-STRs (DYS1003, DYS1007, DYR88, DYS713, DYS712, DYS711, DYS724, DYF1002, DYF1001, DYF1000, DYS1012, DYS1010), which differentiate 26% (424/1616) of father–son pairs. Notably, for all 26 currently known RM Y-STRs, 44% (711/1616) of father‒son pairs were separated26. This finding indicates that increasing the number of RM Y-STR loci significantly enhances the ability to differentiate male relatives. Further studies confirmed that the 26 RM Y-STRs are effective in distinguishing between father–son pairs27,28,29,30.

China is a vast, multiethnic country with complex terrain and a diverse climate. In recent years, Chinese researchers have explored the use of 14 RM Y-STRs within domestic populations22,23,24,25. The Han ethnic group is the largest and most geographically widespread in China. There is a study on 26 RM Y-STRs in Chinese Han males29. However, the available data are insufficient to meet the demands of practical applications. In this study, we explored the mutation rates of 26 RM Y-STRs by analyzing 367 Chinese Han father‒son pairs. Additionally, we assessed the differentiation ability of 26 RM Y-STRs among male relatives with 1 to 4 meioses and unrelated males. Furthermore, all samples were genotyped using the Y-STR kit Y41SE-v1.2, which includes 6 RM Y-STR loci (DYS627, DYS576, DYF387S1, DYS518, DYS570, DYS449) and 30 Y-STR loci (DYS481, DYS389I, DYS635, DYS389II, DYS533, DYS527, DYS460, DYS458, DYS19, DYS444, DYS385, DYS593, DYS393, DYS549, DYS439, DYS392, DYS448, DYS522, DYS456, DYS390, DYS438, Y-GATA-H4, DYS645, DYS459, DYS437, DYS447, DYS391, DYS643, DYS557, DYS596). The latter 30 Y-STR loci are referred to as Y41SE-v1.2 (30 Y-STRs). Y41SE-v1.2 contains all the loci in Yfiler Plus (25 Y-STRs). We compared the differentiation abilities of the 26 RM Y-STRs, Y41SE-v1.2 (30 Y-STRs), and Yfiler Plus in both related and unrelated males. The purpose of this study was to provide empirical evidence supporting the application of RM Y-STRs in forensic medicine.

Materials and methods

Sample collection and DNA extraction

After written informed consent was obtained from each volunteer, blood samples were collected via FTA blood collection cards for this study. A total of 744 male volunteers, representing 289 Chinese Han families, were included, with each family consisting of 2 to 11 males. In terms of familial relationships, among the participants, there were 289 unrelated males, 367 father‒son pairs, 188 pairs with two meioses (117 brothers and 71 grandfather‒grandson pairs), 93 pairs with three meioses (3 great-grandfather‒grandson pairs and 90 uncle‒nephew pairs), and 53 pairs with four meioses (36 cousins and 17 uncle‒great-nephew pairs). If a father had multiple sons, each father–son pair was treated as an independent pair. A Microreader™ 21 Direct ID System Kit (Suzhou YW Gene Technology Co., Ltd., China) was used to analyze and confirm the father‒son relationships. Genomic DNA was extracted using the Chelex-100 method31 and stored at -20 °C for future use. This study was granted approval by the Ethics Committee of North Sichuan Medical College (approval number: 2025005) and was conducted in accordance with the principles of the Helsinki Declaration.

Amplification via the Y41SE-v1.2 kit

PCR amplification was performed according to the instructions of the Y41SE-v1.2 kit. The 25 µL reaction system consisted of 5 µL of Y41SE primers, 10 µL of 2.5× PCR mix, 1 µL of Taq DNA polymerase, and a DNA template from an FTA blood collection card with a diameter of approximately 1.2 mm. The final volume was adjusted to 25 µL with ddH₂O. The amplification conditions were as follows: 95 °C for 3 min; 30 cycles of 94 °C for 5 s, 61 °C for 90 s, and 60 °C for 10 min; and storage at 4 °C.

Amplification of 20 RM Y-STRs

Because the Y41SE-v1.2 kit already includes 6 RM Y-STR loci, the remaining 20 RM Y-STR loci were amplified in separate groups. Sixteen loci were amplified using two multiplex amplification systems: RM1 (DYF404S1, DYF399S1, DYS547, DYS526a/b, DYS626, DYF403S1a/b, DYS612) and RM2 (DYS1003, DYS1007, DYS712, DYS711, DYS724, DYF1002, DYR88, DYF1000). Owing to the poor multiplex PCR amplification efficiency observed at four loci (DYS713, DYS1012, DYF1001, and DYS1010), single-locus amplification was subsequently adopted for these loci. The PCR sequences of primers used were previously described18,26,32 and are shown in Supplementary Table S1. Both RM1 and RM2 used a 10 µL reaction system, which consisted of 2.5 µL of primer mixture, 5 µL of 2×QIAGEN Multiplex PCR Master Mix (Qiagen, Germany), 1.0 µL of DNA template, and 1.5 µL of ddH2O. The amplification conditions were as follows: for RM1, 95 °C for 15 min; 30 cycles of 94 °C for 30 s, 56 °C for 45 s, and 72 °C for 60 s; 60 °C for 45 min; and storage at 4 °C. For RM2, 95 °C for 15 min; 32 cycles of 94 °C for 30 s, 59 °C for 45 s, and 72 °C for 60 s; 60 °C for 30 min; and storage at 4 °C.

Genotyping

The PCR products were analyzed via capillary electrophoresis on an ABI 3500 Genetic Analyzer (Applied Biosystems) with a 36-cm capillary and POP-4 polymer. The data analysis was conducted using GeneMapper® ID-X software with custom panels and bin sets. A detection threshold of 150 RFU was used for peak calling. For the genotypes of RM Y-STRs that have high stutter or varying copy numbers, refer to Ralf et al.32. For the mutated alleles, reamplifications were conducted to confirm the genotype.

Quality control

During the amplification and genotyping processes of the samples, DNA 9948 (Tsingke Biotechnology Co., Ltd., China, 1 ng/µL) and DNase/RNase-free water were used as positive and negative controls, respectively.

Data analysis

Direct counting methods were used to calculate haplotype counts, mutation counts, etc. Allele frequencies were calculated by dividing the number of a specific allele by the total number of alleles at that locus. The discrimination capacity (DC) was calculated using the formula #H/N, where #H is the total number of observed haplotypes and N is the total number of samples. Mutation rates per RM Y-STR were calculated by dividing the number of observed mutations by the number of father–son pairs. The 95% confidence interval (CI) was calculated with Clopper‒Pearson intervals and a binomial distribution (http://statpages.info/confint.html). The significance of the mutation rate per RM Y-STR between populations was assessed via Fisher’s exact test.

Ethical approval

All procedures performed in studies involving human participants followed the ethical standards of the Ethics Committee of North Sichuan Medical College (approval number: 2025005) and complied with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Written informed consent was obtained from all participants and/or their legal guardians included in the study.

Results and discussion

Mutation analysis

A total of 224 mutation events were observed in 367 father–son pairs. Specifically, 174 mutations were observed in 26 RM Y-STRs, 61 were observed in Yfiler Plus, and 50 were observed in Y41SE-v1.2 (30 Y-STRs). There were 215 one-step mutations (96%) and 9 multistep mutations (4%), which is consistent with previous studies by Ralf27 and Lin33. The number of one-step mutations significantly exceeds the number of multistep mutations, which aligns with the classical stepwise mutation model (SMM)34. Among the mutation events, we observed slightly more repeat contractions. There were 126 (56%) repeat expansions and 98 (44%) repeat contractions. However, mutation-induced repeat changes show no preferential expansion/contraction tendency22,23,26,27. Overall, the numbers of expansions and contractions were similar, which is in line with previous research27. The genotyping results for all father–son pairs with observed mutations are provided in Supplementary Table S2.

Table 1 Mutation information of 56 Y-STRs in the Chinese Han population.
Full size table

The number of observed mutations per locus ranged from 0 to 17, with mutation rates varying from < 2.7 × 10− 3 (95% CI: 1.00 × 10− 4‒15.1 × 10− 3) for DYS1007 and DYS626 to 46.7 × 10− 3 (95% CI: 27.2 × 10− 3‒73.1 × 10− 3) for DYS712. The average mutation rate was 18.2 × 10− 3 (95% CI: 15.6 × 10− 3‒21.1 × 10− 3) (Table 1). The Y-STR marker that presented the most mutations was DYS712, with 17 mutations. The mutation rates of six RM Y-STRs (DYS1007, DYS626, DYF403S1b, DYS518, DYS570, DYS449) were lower than 10− 2. No mutations were observed at DYS1007 or DYS626, which differs from previous studies in Chinese Han populations29.

To assess the differences in mutation rates among the 26 RM Y-STRs, we combined the data from this study with previously published data from both Chinese and international populations (Supplementary Table S3). In the combined Chinese Han population, the number of meioses per locus ranged from 674 to 4406, and the average mutation rate varied from 4.5 × 10− 3 (95% CI: 9.00 × 10− 4‒13.0 × 10− 3) for DYS1007 to 54.5 × 10− 3 (95% CI: 48.0 × 10− 3‒61.6 × 10− 3) for DYF399S1. In the combined international population, the number of meioses per locus ranged from 2863 to 5088, and the average mutation rate ranged from 10.0 × 10− 3 (95% CI: 7.40 × 10− 3‒13.6 × 10− 3) for DYF403S1b to 73.6 × 10− 3 (95% CI: 66.4 × 10− 3‒82.2 × 10− 3) for DYF399S1. Among the combined Chinese populations, DYF399S1 had the highest mutation rate. However, in this study, the mutation rate of DYF399S1 was below average, with only 3.27% (12/367) of father–son pairs showing mutations (Table 1). Fisher’s exact test for the 26 RM Y-STRs revealed that the mutation rates of DYS1007 (P value 0.0168), DYF399S1 (P value 0.0002), DYS570 (P value 0.0170), DYS612 (P value 0.0028), and DYS526b (P value 0.0377) were significantly lower in the combined Chinese Han population than in the combined international population.

Additionally, for the 26 RM Y-STRs, we performed Fisher’s exact tests for each pair of mutation events per locus between our study group and other reference populations (Supplementary Table 4). Significant differences in mutation rates were observed at three loci between the Chinese population and the Austrian population: DYS724 (P value 0.0402), DYS1007 (P value 0.0021), and DYF399S1 (P value 0.0058), all of which presented lower mutation rates in our study group27. The mutation rates of DYF399S1 (P value 0.0012) and DYF1001 (P value 0.0029) were significantly lower in our study group than in the Japanese population28. Additionally, significant differences in mutation rates were observed at four loci between the Chinese Han population and the Korean population: DYS712 (P value 0.0013), DYS1007 (P value 0.0458), DYF1001 (P value 0.0398), and DYF399S1 (P value 0.0171), all of which were lower in our study group30.

Moreover, our study revealed that the mutation rates of DYS458, DYS385, and DYS444 were 10.9 × 10− 3, whereas the mutation rates of DYS439 and DYS447 were 13.6 × 10− 3 (Table 1). A significantly higher mutation rate of DYS439 was observed in our study group than in the population studied by Wang et al. (P value 0.0157)15. The mutation rate of DYS458 was lower in our study group than in the Japanese population (33.7 × 10− 3, P value 0.1896). A previous study reported that the mutation rate of DYS385 was higher than 10− 233.

Differentiating related males

We are the first to investigate the ability of 26 RM Y-STRs to differentiate 1–4 meioses among Chinese Han males, as illustrated in Fig. 1. Among the 367 Chinese Han father‒son pairs, Yfiler Plus differentiated only 59 pairs (16.1%), which is higher than reported rates in Austrian (13.4%)27 and Japanese (14.0%) populations28. Y41SE-v1.2 (30 Y-STRs) differentiated only 49 pairs (13.4%). Although Y41SE-v1.2 includes more loci than Yfiler Plus, the addition of six RM Y-STRs in Yfiler Plus increases the likelihood of observing mutations, resulting in a higher differentiation rate for Yfiler Plus. Compared with Yfiler Plus, 26 RM Y-STRs differentiated 134 father‒son pairs (36.5%), which was a significantly greater percentage but was still lower than the 44% differentiation rate reported in previous studies26. When the 26 RM Y-STRs and Y41SE-v1.2 (30 Y-STRs) were combined, 56 Y-STR loci differentiated 44.7% (164/367) of the father‒son pairs. Compared with the 26 RM Y-STRs, only an 8.2% increase in the father–son differentiation rate was observed, demonstrating that adding moderately mutated Y-STR loci provides limited improvement in male relative discrimination.

Fig. 1
Fig. 1
Full size image

Differentiation rates of 1‒4 meioses among male relatives using Y41SE-v1.2 (30 Y-STRs), Yfiler Plus (25 Y-STRs), 26 RM Y-STRs, or Combined (56 Y-STRs). The error bars represent the 95% CIs.

Among the 188 pairs with two meioses, the differentiation rates of Yfiler Plus, Y41SE-v1.2 (30 Y-STRs), 26 RM Y-STRs, and 56 Y-STR loci were 26.6%, 21.3%, 55.9%, and 64.9%, respectively (Fig. 1). Specifically, among the 117 brother pairs, the differentiation rates of Yfiler Plus, Y41SE-v1.2 (30 Y-STRs), and 26 RM Y-STRs were 24.8%, 17.9%, and 55.6%, respectively. Among the 71 grandfather–grandson pairs, the differentiation rates were 26.7%, 29.6%, and 56.3%, respectively. Among the 93 pairs with three meioses, the differentiation rates of Yfiler Plus, Y41SE-v1.2 (30 Y-STRs), 26 RM Y-STRs, and 56 Y-STR loci were 45.2%, 35.5%, 74.2%, and 87.1%, respectively (Fig. 1). In 53 pairs with four meioses, the differentiation rates of Yfiler Plus, Y41SE-v1.2 (30 Y-STRs), 26 RM Y-STRs, and 56 Y-STR loci were 60.4%, 35.8%, 79.2%, and 92.5%, respectively (Fig. 1). Similarly, across 2–4 meioses, the 26 RM Y-STRs demonstrated superior differentiation power compared with both Yfiler Plus and Y41SE-v1.2 (30 Y-STRs). However, all the differentiation rates were lower than those reported in previous studies by Ralf et al.26, who reported differentiation rates of 69%, 83%, and 90%, respectively.

We also investigated the number of Y-STR loci that showed mutational differences between any given differentiation of 1–4 meioses. We found that among the 49 father‒son pairs differentiated by Y41SE-v1.2 (30 Y-STRs), 48 pairs (98.0%) were separated by a mutation in a single Y-STR locus, and 1 pair (2.0%) was separated by mutations in two loci. Among the 59 father‒son pairs differentiated by Yfiler Plus, 57 pairs (96.6%) had only one Y-STR locus mutated, and 2 pairs (3.4%) had mutations in two Y-STR loci. Among the 134 father‒son pairs differentiated by the 26 RM Y-STRs, 106 pairs (79.1%) were separated by a mutation in a single Y-STR locus, 22 pairs (16.4%) were separated by mutations in two loci, 2 pairs (1.5%) were separated by mutations in three loci, 3 pairs (2.2%) were separated by mutations in four loci, and 1 pair (0.74%) was separated by mutations in five loci. The frequency of multiple Y-STR locus mutations in father‒son pairs differentiated by the 26 RM Y-STRs was greater than that observed with Yfiler Plus or Y41SE-v1.2 (30 Y-STRs). Among the father–son pairs differentiated by 56 Y-STRs, there was a notable increase in the number of pairs with mutations in two loci (Fig. 2). These findings demonstrate that both the panel size (number of Y-STR loci) and locus mutation rates constitute critical parameters for optimizing Y-STR-based paternity testing.

Compared with father–son pairs, male relatives with 2–4 meioses are more likely to be differentiated by mutations in multiple Y-STR loci (Fig. 2). This may be due to the increased number of gene transmissions during multiple meioses, which increases the probability of mutations occurring in the loci. Compared with the father–son pairs, five locus mutations were detected in two of the three-meioses pairs using 26 RM Y-STRs, whereas no more than four locus mutations were detected in the two- and four-meioses groups. This may be attributed to the smaller sample size for the 2–4-meioses groups. Among the 56 Y-STRs, seven loci were mutated in one three-meioses pair, and six loci were mutated in one four-meioses pair. Compared with Yfiler Plus and Y41SE-v1.2 (30 Y-STRs), the 26 RM Y-STRs presented higher frequencies of multilocus mutations across 1–4 meiotic divisions (Fig. 2).

Fig. 2
Fig. 2
Full size image

Number of mutations of Y-STR loci in male relatives with 1–4 meioses, analyzed using Y41SE-v1.2 (30 Y-STRs), Yfiler Plus (25 Y-STRs), 26 RM Y-STRs, or the combination (56 Y-STRs).

Differentiating unrelated males

In this study, 289 unrelated males (the eldest male in each family) were selected to obtain their Y-STR haplotypes. Each individual exhibited a unique haplotype across the 56 Y-STRs. For Y41SE-v1.2 (30 Y-STRs), 285 unique haplotypes were obtained from the 289 unrelated males; 281 haplotypes were found in single males, and 4 haplotypes were shared between two males, resulting in a haplotype discrimination capacity of 98.6%. For Yfiler Plus, 288 unique haplotypes were obtained from the 289 unrelated males, with 287 haplotypes found in single males and 1 haplotype shared between two males, resulting in a haplotype discrimination capacity of 99.7%. For the 26 RM Y-STRs, each individual carried a unique haplotype, resulting in a haplotype discrimination capacity of 100%. These findings suggest that RM Y-STRs enhance the ability to differentiate unrelated males, which differs slightly from previous studies based on an Austrian population27. The relatively small sample size in this study may limit the generalizability of these results, and larger sample sizes will be necessary in future studies to obtain more reliable diversity estimates. Additionally, 33 abnormal genotypes were observed among the 56 Y-STR loci for 289 unrelated males, with 27 occurring in the multicopy RM Y-STRs. For example, 137 F and 137 S presented a 3-copy genotype at DYS724 (reported as a two-copy locus) and a 4-copy genotype at DYF399S1 (reported as a three-copy locus), which has also been described in previous studies26,29. Compared with the autosomal chromosome, the Y chromosome harbors fewer genes with lower density, resulting in reduced selective pressure that facilitates large-scale insertion events. Furthermore, the absence of recombination on the Y chromosome allows cumulative retention of these insertions35. The above factors may explain the appearance of multicopy STR loci on the Y chromosome and the susceptibility to abnormal typing. For abnormal genotypes, we can test genotypes again, or perform sequencing, or test their male child, but there may be differences in the number of alleles between fathers and sons; for example, 244 F showed a 4-copy genotype at DYF399S1, 244 S showed a 3-copy genotype, and it is calculated as a mutation (Table S2). The presence of abnormal allele typing in Y-STRs can provide insights into paternal locus characteristics, which may increase paternal identification rates and guide case investigations. The allele frequencies observed in this study are provided in Supplementary Table S5.

Conclusion

In this study, we investigated the mutation rates of 56 Y-STRs in 367 Chinese Han father‒son pairs. Six RM Y-STRs (DYS1007, DYS626, DYF403S1b, DYS518, DYS570, and DYS449) presented mutation rates lower than 10− 2, whereas the mutation rates of DYS458, DYS385, DYS439, and DYS447 were higher than 10− 2. Furthermore, a comparison of the mutation rates of the 26 RM Y-STRs between Chinese and international reference populations revealed significant differences at five loci. This study also provides the first report on the ability of 26 RM Y-STRs to differentiate between 1 and 4 meioses in the Chinese Han population. Compared with Yfiler Plus and Y41SE-v1.2 (30 Y-STRs), the 26 RM Y-STRs significantly improved the differentiation rate for related males. Additionally, the 26 RM Y-STRs demonstrated distinct advantages in differentiating unrelated males compared with Yfiler Plus and Y41SE-v1.2 (30 Y-STRs). In summary, the 26 RM Y-STRs exhibit strong potential for forensic applications, particularly in individual identification and paternal kinship testing among Chinese Han populations. Future research should further explore populations with diverse genetic backgrounds to increase the utility of this analysis in forensic practice for identifying both related and unrelated males.