Search for: All records

Abstract Dispersal can evolve as an adaptation to escape competition with conspecifics or kin. Locations with a low density of conspecifics, however, may also lead to reduced opportunities for mating, especially in sessile marine invertebrates with proximity-dependent mating success. Since there are few experimental investigations, we performed a series of field experiments using an experimentally tractable species (the bryozoan Bugula neritina) to test the hypothesis that the density, spatial arrangement, and genetic relatedness of neighbours differentially affect survival, growth, reproduction, paternity, and sperm dispersal. We manipulated the density and relatedness of neighbours and found that increased density reduced survival but not growth rate, and that there was no effect of relatedness on survival, growth, or fecundity, in contrast to previous studies. We also manipulated the distances to the nearest neighbour and used genetic markers to assign paternity within known mother–offspring groups to estimate how proximity affects mating success. Distance to the nearest neighbour did not affect the number of settlers produced, the paternity share, or the degree of multiple paternity. Overall, larger than expected sperm dispersal led to high multiple paternity, regardless of the distance to the nearest neighbour. Our results have important implications for understanding selection on dispersal distance: in this system, there are few disadvantages to the limited larval dispersal that does occur and limited advantages for larvae to disperse further than a few 10s of metres. 

Many and varied methods currently exist for featurization, which is the process of mapping persistence diagrams to Euclidean space, with the goal of maximally preserving structure. However, and to our knowledge, there are presently no methodical comparisons of existing approaches, nor a standardized collection of test data sets. This paper provides a comparative study of several such methods. In particular, we review, evaluate, and compare the stable multi-scale kernel, persistence landscapes, persistence images, the ring of algebraic functions, template functions, and adaptive template systems. Using these approaches for feature extraction, we apply and compare popular machine learning methods on five data sets: MNIST, Shape retrieval of non-rigid 3D Human Models (SHREC14), extracts from the Protein Classification Benchmark Collection (Protein), MPEG7 shape matching, and HAM10000 skin lesion data set. These data sets are commonly used in the above methods for featurization, and we use them to evaluate predictive utility in real-world applications.