View article

DNA recovered from non-invasive samples, in particular faeces, is usually degraded and/or in small quantities. This often leads to PCR errors during microsatellite genotyping (due to allelic dropout or false alleles), resulting in the identification of incorrect genotypes (Taberlet et al. 1996, 1999; Smith et al. 2000). This is a major concern because microsatellite genotypes are commonly used for individual identification, parentage, relatedness, and population genetics (eg, Taberlet et al. 1997; Constable et al. 2001; Garnier et al. 2001). PCR methods to overcome genotyping errors when the template concentration is low have to date relied on replications (eg, Navidi et al. 1992; Taberlet et al. 1997, 1999; Morin et al. 2001). However, high degrees of replication can exhaust a sample very quickly as well as increasing the consumables, costs and time required (Morin et al. 2001). To avoid the need for numerous replicates, and to maximise the proportion of faecal samples that contains the critical threshold amount of DNA for accurate genotyping (Taberlet et al. 1996; Morin et al. 2001), we devised a two-step PCR method with the aim of increasing the quality and quantity of the desired DNA template. The basis of this approach is that an initial large-volume (50 ll) PCR is carried out containing primers for the entire panel of loci to be genotyped. Two microlitres of the amplified product from this first PCR are then used as template in separate PCRs for each locus. We compared this new method with conventional single-step PCR on the same faecal DNA extracts from three Australian marsupials (the brush-tailed rock-wallaby, Petrogale penicillata; the spotted …