Using a multistate occupancy approach to determine molecular diagnostic accuracy and factors affecting avian haemosporidian infections.

The use of a sensitive and accurate parasite detection methodology is crucial in studies exploring prevalence of parasites in host populations or communities, and uncertainty in identifying parasite genera and/or lineages may limit the understanding of host-parasite interactions. Here, we used a multistate occupancy approach that accounts for imperfect detection to assess whether sex and breeding season influenced the prevalence of a specific Haemoproteus lineage (TARUF02) in a white-lined tanager population. Likewise, we explored whether the probability of detecting the target parasite in an infected bird using PCR and sequencing analyses may be influenced by season and host sex. We found little evidence that sex influenced the probability of an individual host being infected by a haemosporidian parasite. Conversely, we found that the probability of infection by Haemoproteus TARUF02 was ~30% higher during the breeding season, reflecting a higher prevalence of this parasite in this season. The probability that PCR detects DNA of haemosporidian parasite was higher for female birds, suggesting that they are more prone to be parasitized with parasitemia levels that are more successfully detected by molecular analysis. Sequencing successfully determined the Haemoproteus TARUF02 lineage in 60% of samples collected during the breeding season and 84% of samples collected during the non-breeding season. Understanding the ecology of hosts and aspects of their physiology that may influence the parasite infection is essential to better understanding of hemoparasite infections and how parasites influence their native hosts, through decreasing reproductive success, lifespan, and/or survival.

ATLANTIC-CAMTRAPS: a dataset of medium and large terrestrial mammal communities in the Atlantic Forest of South America.

Our understanding of mammal ecology has always been hindered by the difficulties of observing species in closed tropical forests. Camera trapping has become a major advance for monitoring terrestrial mammals in biodiversity rich ecosystems. Here we compiled one of the largest datasets of inventories of terrestrial mammal communities for the Neotropical region based on camera trapping studies. The dataset comprises 170 surveys of medium to large terrestrial mammals using camera traps conducted in 144 areas by 74 studies, covering six vegetation types of tropical and subtropical Atlantic Forest of South America (Brazil and Argentina), and present data on species composition and richness. The complete dataset comprises 53,438 independent records of 83 species of mammals, includes 10 species of marsupials, 15 rodents, 20 carnivores, eight ungulates and six armadillos. Species richness averaged 13 species (±6.07 SD) per site. Only six species occurred in more than 50% of the sites: the domestic dog Canis familiaris, crab-eating fox Cerdocyon thous, tayra Eira barbara, south American coati Nasua nasua, crab-eating raccoon Procyon cancrivorus and the nine-banded armadillo Dasypus novemcinctus. The information contained in this dataset can be used to understand macroecological patterns of biodiversity, community, and population structure, but also to evaluate the ecological consequences of fragmentation, defaunation, and trophic interactions.