Mutation rates of 22 autosomal STR loci in a European population from central Balkan, Republic of Serbia

Abstract

Analysis of short tandem repeats (STRs) is a tool for human DNA identification, including paternity and criminal investigations. Locus-specific STR mutation rates are critical for interpretation of DNA identification in practice. Accumulating evidence suggests that STR mutation rates are population specific. However, STR mutation rates based on trios have not been analyzed in European population yet. In this study, mutation rates of 20 CODIS and two additional autosomal STR loci (Penta E and Penta D) were determined from 1279 cases of paternity testing in the Serbian population, and the relationships between STR mutation rates and sex, allele length, and heterozygosity were investigated. A total of 63 mutations were observed at 18 loci in a total of 28278 allele transmissions, including 62 (98.4%) single-step and 1 (1.6%) two-step mutations. The average mutation rate was 1.7×10-3 (95% CI: 1.3-2.3×10-3), whereas locus-specific mutation rates ranged from 5.6×10-3 (D12S391) to 0.6×10-3 (D2S1338, D5S818, and D21S11). No mutation was observed at the TPOX, TH01, D16S539, and D22S1045 loci. The mutation rate of paternal origin (2.4×10-3) was eight times higher than that of maternal origin (0.3×10-3), and long alleles mutated more frequently than short and medium alleles (χ2=4.436 and χ2=4.646, respectively, p<0.05). No differences were found when analyzing overall expansion versus contraction (χ2=0.999, p=0.317). Among short alleles two expansions and no contraction were detected, while among long ones three expansions and nine contractions were observed. Furthermore, we found no correlation between locus-specific mutation rates and corresponding heterozygosity (rs=-0.188, p=0.559). Comparisons between populations revealed statistically significant differences in locus-specific mutation rates of 13 CODIS STRs between Serbian population and Chinese or Brazilian. Our results show that STR mutation rates depend on sex, allele size and population, and provide useful data on STR mutation rate based on trios for human DNA identification in European populations.