RESUMEN
We have been testing familial relationships based on short tandem repeats (STRs) in families who requested it either voluntarily or by order of the court. Here, we present a summary of our 5-year experience of autosomal STR-based paternity tests. A total of 1,431 individuals from 588 cases were tested, including 878 pairs of either of the parent, and a child. Among these 588 cases, genetic information about the other parent was available only for 135 cases. Five hundred eighteen pairs were concluded to be parent-child relations, for which the median paternity index (PI) was 72,826, and the median decimal logarithm was 4.860. Autosomal mutation was observed in nine pairs (1.74%), and the pairs harbored only one mismatched locus among the 15 standard loci (D8S1179, D21S11, D7S820, CSF1PO, D3S1358, TH01, D13S317, D16S539, D2S1338, D19S433, vWA, TPOX, D18S51, D5S818, and FGA). The number of mismatched loci did not increase even after additional loci were included in the study. The observed mutation rates were D13S317 (0.193%), D18S51 (0.193%), D19S433 (0.193%), FGA (0.193%), vWA (0.386%), Penta D (0.387%), and Penta E (0.193%). There were 14 pairs with two mismatched loci, which we excluded through additional tests on either autosomal or X chromosomal STRs, and mitochondrial sequencing. Although PI is useful for determining parent-child relation, it provides indirect information; it is an interpretation of the test results that is based on probability. Additional genotyping on sex chromosome and mitochondrial DNA, or participation of other family members might be beneficial for a reliable conclusion.