Misidentification error in non-invasive genetic mark-recapture sampling

Abstract

Advances within the fields of molecular and genetic biology have increased the ability to use genetic analyses in wildlife studies, especially with non-invasive sampling methods. The presence of genetic error, namely allelic dropout and false alleles, can bias demographic estimates from non-invasive studies. The balance of accepting certain levels of error versus genotyping additional loci has not been quantified in the literature. The main objective of this study was to develop a cost-effective method of estimating misidentification error from noninvasive sampling. Error could then be incorporated into black bear abundance estimates for a case study in the central Georgia population (CGP). Calibration samples of known individuals from tissue and hair samples collected from the CGP were used. Model verification was conducted with both simulated data and data from the case study. The objective function for optimal selection of a marker panel with and without genetic error was also described.