Measuring Similarity among Hispid Cotton Rats (Sigmodon hispidus) of Known Relatedness with DNA Fingerprinting

Abstract

Determining mating strategies and social behavior of natural populations of mammals is equivocal when individuals can not be observed directly. DNA fingerprinting holds promise to elucidate relationships among individuals indirectly; however, few empirical data assess accuracy of DNA fingerprints in portraying relatedness in natural populations of mammals. Using the hispid cotton rat ( Sigmodon hispidas ) as a model, we determined accuracy of DNA fingerprinting for separating wild-caught individuals into categories according to their relatedness and ascertained repeatability of the pattern of DNA fingerprints. We used individuals from known breedings to compute similarity (S) in DNA fingerprints from pair-wise matchings. Because relatedness of these individuals was known, we obtained the actual distribution of band-sharing scores and computed the known mean S for each relatedness category. We detected significant differences in S between primary, secondary, and unrelated categories but could not distinguish statistically among related groups within these categories. High within-gel repeatability in the DNA banding pattern indicated a robust technique, but variation in scoring among readers confirmed that a single individual should score gels. Our results demonstrated that DNA fingerprints can be used to determine relatedness categories and uncover behavioral-social interactions for species that can not be observed directly. Routine demographic information (e.g., body mass, spatial proximity) collected during field studies and used in conjunction with DNA fingerprinting can provide a finer level of distinction among related groups.