%0 Generic %A Dallas, Tad %D 2016 %T Data and code to reproduce Dallas, Park, and Drake 2016 "Predictability of helminth parasite host range using information on geography, host traits and parasite community structure" %U https://figshare.com/articles/dataset/Data_and_code_to_reproduce_Dallas_Park_and_Drake_2016_Predictability_of_helminth_parasite_host_range_using_information_on_geography_host_traits_and_parasite_community_structure_/3795330 %R 10.6084/m9.figshare.3795330.v1 %2 https://ndownloader.figshare.com/files/5909352 %2 https://ndownloader.figshare.com/files/5909355 %K Helminth %K FishPest %K Host-parasite interactions %K Species distribution model %K Boosted regression tree %K Parasite niche %K Host-Parasite Interactions %X This repository contains files necessary to reproduce analyses and figures from 

Dallas, T., A.W. Park, and J.M. Drake. 2016. Predictability of helminth parasite host range using information on geography, host traits and parasite community structure. Parasitology. doi:10.1017/S0031182016001608. 

The aim of the paper was to predict the set of host species capable of being infected by a given parasite. Models were trained for each parasite species in a set of over 500 helminth parasites. We found that the existing parasite community was important to accurate identification of permissive host species, suggesting that the parasite community of a host species contains information that is more useful for predicting host suitability than host traits, geographic location, or host taxonomy. 

Note: the code currently requires a minimum of a 5 core workstation (the n.cores argument can be changed though in the gbm function). Be mindful of memory usage as well. While I don't recall exact benchmarks, I believe the full set of models took at least 48 hours to run on a decently equipped workstation (16 core; 3.0 Ghz processor). 



%I figshare