We performed a population study on 15 polymorphic STR loci (FH2010, FH2079, PEZ2, VWF.X, FH2054, FH2087Ub, FH2611, WILMS-TF, PEZ12, PEZ15, PEZ6, FH2087Ua, ZUBECA4, ZUBECA6, FH2132) on 131 randomly selected dogs. Alleles were identified and grouped according to their estimated fragment length using fixed allelic bins encompassing one base-pair. The allele assignment was confirmed by sequence analysis of homozygote and cloned heterozygote alleles. In order to develop a uniform repeat-based nomenclature, extensive sequence analysis was performed on a selection of alleles from each STR locus. The proposed nomenclature refers to the internationally recognised recommendations for human-specific STR loci in forensic applications. The 15 canine-specific STR loci were grouped into 3 classes (simple STRs, compound STRs and complex/hypervariable STRs) according to their complexity and variability within the repeat structure. Finally, we evaluated the precision of fragment size estimation on a capillary electrophoresis platform and demonstrated reproducibility of fragment length estimation for single base-pair intermediate alleles.

The traditional protocol for forensic mitochondrial DNA (mtDNA) analyses involves the amplification and sequencing of the two hypervariable segments HVS-I and HVS-II of the mtDNA control region. The primers usually span fragment sizes of 300-400 bp each region, which may result in weak or failed amplification in highly degraded samples. Here we introduce an improved and more stable approach using shortened amplicons in the fragment range between 144 and 237 bp. Ten such amplicons were required to produce overlapping fragments that cover the entire human mtDNA control region. These were co-amplified in two multiplex polymerase chain reactions and sequenced with the individual amplification primers. The primers were carefully selected to minimize binding on homoplasic and haplogroup-specific sites that would otherwise result in loss of amplification due to mis-priming. The multiplexes have successfully been applied to ancient and forensic samples such as bones and teeth that showed a high degree of degradation.

The canine mitochondrial DNA (mtDNA) control region of 133 dogs living in the area around Innsbruck, Austria was sequenced. A total of 40 polymorphic sites were observed in the first hypervariable segment and 15 in the second, which resulted in the differentiation of 40 distinct haplotypes. We observed five nucleotide positions that were highly polymorphic within different haplogroups, and they represent good candidates for mtDNA screening. We found five point heteroplasmic positions; all located in HVS-I and a polythymine region in HVS-II, the latter often being associated with length heteroplasmy. In contrast to human mtDNA, the canine control region contains a hypervariable 10 nucleotide repeat region, which is located between the two hypervariable regions. In our population sample, we observed eight different repeat types, which we characterized by direct sequencing and fragment length analysis. The discrimination power of the canine mtDNA control region was 0.93, not taking the polymorphic repeat region into consideration.

In this study a proposal for the allele nomenclature of six polymorphic short tandem repeat (STR) loci (PEZ3, PEZ6, PEZ8, PEZ10, FHC2161, and FHC2328) for canine genotyping (Canis lupus familiaris) is presented. The nomenclature is based on the sequence data of the polymorphic region of the microsatellite markers as recommended by the DNA commission of the International Society of Forensic Haemogenetics (ISFH) in 1994 for human DNA typing. To cover commonly and rarely occurring alleles, a selection of homozygous and heterozygous animals were analyzed and subjected to sequence studies. The alleles consisted of simple tri- and tetra-nucleotide repeat patterns as well as compound and highly complex repeat patterns. Several alleles revealing the same fragment size but different repeat structures were found. The allele designation described here was adopted to the number of repeats, including all variable regions within the amplified fragment. In a second step the most commonly occurring alleles were added to an allelic ladder for each marker allowing a reliable typing of all alleles differing in size. A total number of 142 unrelated dogs from surrounding municipal animal homes, private households, and canines in police duty were analyzed. The data were added to a population database providing allele frequencies for each marker.

Forensic investigations in dog attacks usually involve the examination of bite marks and toothprints, the dog's stomach and pathological methods. For identification of the offending dog we evaluated canine STR typing of saliva traces on dog bite marks. The specificity of 15 canine-specific STRs was tested on human-canine DNA mixtures prior to an applied study in which 52 cases of dog bites were investigated. The first-aid wound bandages as well as swab samples from the surrounding area of the wound were used for DNA analyses. Generally, it was possible to obtain a canine-specific STR profile from the dog's saliva left on the wound area, even when high background of human DNA was present (blood). Interestingly, we found canine STR typing to be more successful when the bandages and swabs showed high amounts of human blood, i.e. when the dog bite was severe. Canine saliva was then sometimes visible as white-coloured secretion on the human blood surface. Less severe bite cases, which did not result in bleeding wounds, showed less success in obtaining useful STR results, probably due to the fact that the surface of the wounds may have been treated before the victims consulted medical aid which therefore removed the canine cells.