Effect of sample age and season of collection on the reliability of microsatellite genotyping of faecal DNA

Individual identification of animals from DNA in field-collected faecal samples is becoming an increasingly important tool in wildlife population monitoring. A major issue relevant to the application of this technique is the reliability of the genotypes obtained. I investigated the effect of sample age and season of collection on amplification rates and reliability of microsatellite genotypes amplified from faecal DNA of a marsupial herbivore, the brush-tailed rock-wallaby (Petrogale penicillata) and a eutherian carnivore, the red fox (Vulpes vulpes). Comparison of DNA profiles from 1 day to 6 months for both species suggests that as the age of the faeces increases there is less good-quality DNA present on the surface of the faeces, resulting in significantly decreasing amplification rates and increasing genotyping error rates over time. No microsatellite PCR products were obtained from samples older than 3 months from any faecal DNA extract in either season. For both species, faeces collected during the summer trial yielded high-quality DNA for up to one week. Faeces collected in winter had significantly lower amplification rates and higher genotyping errors than the summer-collected samples. Computer simulations were used to estimate the probability of obtaining false genotypes when genotyping faecal samples of various ages. These revealed that three replicates is sufficient to prevent identification of false individuals for P. penicillata from faeces up to one week old in both summer and winter but more replicates may be required for older samples, particularly in winter. In contrast, up to eight replicates may be required for fox faeces collected in winter, particularly if more than one week old. These results also suggest that it is difficult to visually identify faecal age for V. vulpes, and any study using fox faeces would need to account for the likely inclusion of older faeces in a field collection. For P. penicillata, faecal age could be accurately assessed, particularly when less than one week old and targeting faeces that match the two most reliable appearance classes described here would be an efficient sampling strategy. It is recommended that the appropriate PCR replication protocol for any given study should be tailored to the error rates expected for the oldest samples likely to be collected. This study is the first to thoroughly investigate the effects of sample age and season of collection on microsatellite genotyping from faecal samples and provides guidelines for sampling and PCR repetition strategies for field-based non-invasive DNA studies.