Introduction

Polymorphisms on the Y chromosome have become a very useful tool in forensic and population genetics. A number of short tandem repeat (STR) polymorphisms have been widely studied in Europe and other areas, and large databases have been made publicly available [1]. Thus, population frequency data of a standard set of STRs combined in a so-called minimal haplotype are now available through the internet database of the Y-user group [1]. Such a haplotype is established by typing 9 loci: DYS19, DYS389I, DYS389II, DYS390, DYS391, DYS392, DYS393, and DYS385I-II. Other STRs have also been recently suggested as interesting for forensic purposes [2, 3, 4]. However, given the linkage among different loci on the non-recombining part of the Y chromosome, it is unclear if they add useful information to the widely studied minimal haplotype, or result in redundant information.

Therefore, we planned to study allele population distribution and forensic power of a recently described set of eight STRs, and compare them with the minimal haplotype.

Materials and methods

Subjects

Unrelated male individuals (n=134) living in Cantabria, a region in northern Spain with a population about 500,000, were studied. Genomic DNA was extracted from peripheral blood by a commercial method according to the manufacturer's instructions (Qiagen, Germany) and quantified by light absorbance (Genequant, Pharmacia, Sweden). Aliquots containing 10–30 ng DNA were used to amplify the regions of interest by PCR, with the primers reported by González-Neira et al. [4], except for loci DYS460 (GATA A7.1), DYS461 (GATA A7.2) and GATA H4. These loci were amplified with new primers designed to decrease the size range of the alleles, as recently reported in the collaborative exercise by the GEP-ISFG (Spanish and Portuguese Group of the International Society of Forensic Genetics) [5, 6] (Table 1). Two multiplex reactions were carried out. Multiplex 1 included FAM-labelled primers to amplify GATA A7.1 (0.3 μM), GATA A10 (0.2 μM), GATA H4 (0.5 μM),and DYS439 (0.2 μM). Multiplex 2 contained TET-labelled primers to amplify GATA A7.2 (0.2 μM), GATA C4 (0.5 μM), DYS437 (0.16 μM), and DYS438 (0.5 μM). Both reaction mixes contained 2 mM MgCl2, 200 μM dNTPs and 0.5 U Taq Gold polymerase (Applied Biosystems, USA) in a 25 μl volume. Cycling conditions for both multiplexes were: pre-incubation at 95° for 7 min, followed by 32 cycles of 30 s at 94°, 20 s at 60° and 30 s at 70°, with a final extension step of 45 min at 70°C.

Table 1 . Primer sequences of the Y-chromosome STRs investigated
Full size table

DYS385 locus was analysed with the protocol proposed by Schneider et al. [7]. Other loci in the minimal haplotype were amplified with the primers described by Gusmão et al. and Kayser et al. [8, 9, 10], as previously reported [11].

Amplicon analysis

Aliquots containing 1.2 μl of PCR product were mixed with 24 μl formamide and 1 μl TAMRA as a internal size standard, heated for 5 min at 95°C, quenched in an ice bath for 10 min and injected into an ABI 310 capillary electrophoresis system (Applied Biosystems). Amplicon size was determined with the local Southern method implemented in Genescan software. Allele designation followed the ISFG and other previously published recommendations [12, 13]. Sequenced alleles were used as controls.

Data analysis

Allele and haplotype frequencies were estimated with Arlequin software (Schneider et al, University of Geneva; available at http://anthro.unige.ch/arlequin). Locus and haplotype diversities were calculated as (1-ΣPi2)(n/n-1), where Pi is the allele or haplotype frequency [14]. The haplotype matching probability was estimated as 1−haplotype diversity.

Results and discussion

Allele frequencies of the loci included in multiplex 1 are shown in Table 2; and those of loci included in multiplex 2 in Table 3. If only the 4 loci included in multiplex 1 were considered, 59 different haplotypes were found among the 134 subjects studied, whereas multiplex 2 determined 41 different haplotypes. Thus, loci included in multiplex 1 were more useful for discrimination purposes. The combined analysis of both multiplexes resulted in 98 different haplotypes: 2 haplotypes were found in 6 individuals, 1 haplotype in 5, 6 in 3, 10 in 2, and 79 in a single individual. Haplotype diversity was 0.973 for multiplex 1, 0.891 for multiplex 2, and 0.992 for the combined 8-locus haplotype.

Table 2 . Allele frequencies of loci included in multiplex 1 (n=134)
Full size table
Table 3 . Allele frequencies of loci included in multiplex 2 (n=134)
Full size table

Allele frequency distributions in this population group were similar to the distribution in other Iberian populations [6, 12], but different from the Oriental ones [15].

A population subset (n=104) was also genotyped for the standard minimal haplotype, in order to compare haplotype diversity of the different allele combinations (population data for some of those loci have been previously published [11]). The results obtained are shown in Tables 4 and 5. A similar haplotype diversity was obtained with the combined two newer multiplexes reported here, and with the standard minimal haplotype. Adding allele information derived from multiplex 1 to the standard minimal haplotype increased discrimination ability, but further addition of multiplex 2 resulted in only a small gain (Table 5).

Table 4. Haplotypes resulting from the analysis of loci included in multiplex 1, multiplex 2 and the minimal haplotype (n=104)
Full size table
Table 5 . Haplotype diversity for different Y-STR sets (n=104 individuals)
Full size table

The lack of recombination and the subsequent linkage among different loci on the Y chromosome means that marginal utility (i.e., the gain of discrimination power attained by including new loci) decreases as more STRs are included into the analysis. The ideal number of loci to study has not been firmly established, and is probably different depending on the purposes of the study and, in forensic cases, on the availability of other genetic data. Certainly, there is some degree of compromise between the power of discrimination attained and the cost and effort of allele typing.

In the present study, the power of discrimination was similar to that found in a recent European study by Bosch et al. who reported a minimal haplotype diversity of 0.9896, which increased to 0.9988 after including data from 10 additional loci [16]. Somewhat larger diversity values were found in Asian populations. Tsai et al. reported a 0.9999 diversity for the minimal haplotype plus the DYS388 locus in Taiwan [17]. Shin et al. found that a combined haplotype including the minimal loci plus DXYS156 and DYS388 displayed a 0.9995 diversity in Korea [18].

Our results suggest that the 9-locus standard minimal haplotype, as defined by the Y-user group [1] and the 8-locus haplotype determined by using multiplexes 1 and 2 reported here have a similar power of discrimination. Both are likely to be discriminative enough in many routine cases. Combining the minimal haplotype with the multiplex 1 (DYS460, GATA A10, GATA H4, DYS 439) into a 13-locus Y-STR set further improves discrimination power, up to a figure similar to that recently reported for the combined analysis of 19 loci [16]. However, little is gained by the further inclusion of loci in multiplex 2.