Degradation Rates of DNA Extracted from Fecal Samples for Use with Molecular Markers
Task Manager:Sara Oyler-McCance
Noninvasive genetic sampling of cougars may prove to be a useful method to estimate population size of cougars across the state, which is something that has not been successfully done using traditional methods. The use of hair is probably not an effective method for felids because of the amount of shedding, which limits the number of samples that actually produce DNA. Scats however, should provide a useful method of collecting genetic material as epithelial cells are shed from the intestinal track and appear in the feces. Important considerations that must be made are the timing of collections and the method of sampling in order to make valid inferences about population size. Defining a sampling scheme for the collection of scats is critical to obtaining unbiased estimates of population size.
Using radio-collared or captured cats and invasively collected genetic samples could be used as a means of identifying cats or producing the initial marking event but resampling must be done independently of knowledge about cat locations. Potentially transects could be searched for cougar scats but this is likely to be inefficient and yield low sample sizes. The use of scat-searching dogs may also be possible but efficiency may still be low, especially in large areas where access may be limited. Another option would be to use hunter harvest as the resampling event. Concerns with this would be sex and age bias in the harvest, but stratification may overcome these issues. A final potential approach would be to mark prey and locate scats from known kill sites. The only potential bias with this would be if marking prey affected how cats selected prey. This approach would benefit from the short time span between scat deposition and collection. Other considerations when using scat are the quantity and quality of genetic material that can be obtained from the sample, which are factors of time since deposition and environmental conditions. Extrinsic factors, such as season, diet, and sample age, can cause variation in genotyping error rates. Potential weaknesses with noninvasive genetic sampling are low success rates, contamination concerns, and high genotyping error rates. Even small genotyping errors (< 1%) cause reasonably large overestimation of population size. Some of these errors include allelic dropout (alleles fail to amplify) and false alleles (spurious amplification of an allele that does not exist for a given individual). In a review of 21 population inventory studies, it has been stated that it is unclear what the frequencies of errors are for different types of errors in relation to the type and quality of the samples being collected and suggests that use of noninvasive sampling will be limited unless evidence is provided to support the reliability of the data. It is also evident that siblings or closely related individuals provide the greatest challenge to genetically distinguishing individuals. A marker system with the power to distinguish a small number of closely related individuals will also have sufficient power to distinguish among a large number of unrelated individuals. Noninvasive genetic sampling of cougar feces is one of the few methods available that may provide realistic estimates of cougar populations in Colorado, which is important since these populations are being exploited. Before genetic sampling can be conducted, an assessment of potential genetic error rates must be made in order to appropriately design the sampling protocols and determine if such a method can provided reliable estimates. Currently there are 3 sibling cougar in captivity at the CDOW Wildlife Health Lab, which can be used to produce known age feces. These scat samples can be subjected to both controlled and uncontrolled environmental conditions in order to accurately assess the magnitude of error rates that may be expected if scats are used for noninvasive population estimation of cougar.
For more information contact:Sara Oyler-McCance