Estimating Differential Reproductive Success From Nests of Related Individuals, With Application to a Study of the Mottled Sculpin,Cottus bairdi

Abstract

Understanding how variation in reproductive success is related to demography is a critical component in understanding the life history of an organism. Parentage analysis using molecular markers can be used to estimate the reproductive success of different groups of individuals in natural populations. Previous models have been developed for cases where offspring are random samples from the population but these models do not account for the presence of full-and half-sibs commonly found in large clutches of many organisms. Here we develop a model for comparing reproductive success among different groups of individuals that explicitly incorporates within-nest relatedness. Inference for the parameters of the model is done in a Bayesian framework, where we sample from the joint posterior of parental assignments and fertility parameters. We use computer simulations to determine how well our model recovers known parameters and investigate how various data collection scenarios (varying the number of nests or the number of offspring) affect the estimates. We then apply our model to compare reproductive success among different age groups of mottled sculpin, Cottus bairdi, from a natural population. We demonstrate that older adults are more likely to contribute to a nest and that females in the older age groups contribute more eggs to a nest than younger individuals.

Parentage and Sibship Inference From Multilocus Genotype Data Under Polygamy

“…Larger family sizes actually render our method more powerful, as shown in Figure 2. There are some claims that large families tend to be split by our method (e.g., Jones et al 2007). However, our analyses of both simulated (100 offspring per full-sib family) and empirical [e.g., 44 $ 47 offspring per full-sib family of an ant data set (Wang 2004)] data sets show that our method recovers large families completely even when a moderate number of six SSRs are used.…”

“…Other methods for comparison: Although many methods are available, few allow for the inference of sibship and parentage jointly (but see Emery et al 2001;Jones et al 2007) or separately in the same framework. The method proposed by Jones et al (2007) has not been implemented in any software ready to use.…”

“…The method proposed by Jones et al (2007) has not been implemented in any software ready to use. The method proposed by Emery et al (2001) is implemented in the software PARENTAGE, but some simulations indicate that it does not converge reliably for the joint sibship and parentage inference except for very small samples.…”

“…Few methods exist for estimating simultaneously parentage and full and half sibship among any number of individuals (Emery et al 2001;Jones et al 2007). This is unfortunate because, on one hand, background relationship interferes with the inference of the focal relationship and thus reduces the statistical power (Wang 2004), and, on the other, taking the background relationship into account by inferring multiple relationships among three or more individuals gains enormously more power (Sieberts et al 2002;Wang 2007).…”

Full-sib reconstruction in haplodiploid populations

“…(1)). Most of these methods are computationally intensive and, in many cases, do not work successfully on challenging data sets in which family contributions were strongly biased (Butler et al, 2004;Jones et al, 2007). Other approaches using fractional assignment with the Bayesian statistics (Almudevar, 2007) or using molecular coancestry (Nomura, 2008) also need large samples or huge number of marker loci.…”