Discrimination of fish populations using parasites: Random Forests on a ‘predictable’ host-parasite system

Abstract

We address the effect of spatial scale and temporal variation on model generality when forming predictive models for fish assignment using a new data mining approach, Random Forests (RF), to variable biological markers (parasite community data). Models were implemented for a fish host-parasite system sampled along the Mediterranean and Atlantic coasts of Spain and were validated using independent datasets. We considered 2 basic classification problems in evaluating the importance of variations in parasite infracommunities for assignment of individual fish to their populations of origin: multiclass (2-5 population models, using 2 seasonal replicates from each of the populations) and 2-class task (using 4 seasonal replicates from 1 Atlantic and 1 Mediterranean population each). The main results are that (i) RF are well suited for multiclass population assignment using parasite communities in non-migratory fish; (ii) RF provide an efficient means for model cross-validation on the baseline data and this allows sample size limitations in parasite tag studies to be tackled effectively; (iii) the performance of RF is dependent on the complexity and spatial extent/configuration of the problem; and (iv) the development of predictive models is strongly influenced by seasonal change and this stresses the importance of both temporal replication and model validation in parasite tagging studies.

Variación geográfica de las comunidades de parásitos de la merluza de tres aletas Micromesistius australis al sur de Sudamérica

“…Otros comparan las comunidades de parásitos (ya sea infracomunidades o comunidades componentes) entre zonas en estudio, sin hacer distinción por el tipo de parásito (e.g., Lester & MacKenzie 2009 para una crítica). Además, entre estos últimos hay variedad de métodos empleados para el análisis de los datos, que motiva discusión (Perdiguero-Alonso et al 2008, Pérez-del-Olmo et al 2010. Ciertamente, los estudios parasitológicos que disponen información adicional de los hospedadores (morfometría, microquímica de otolitos, genética) son más completos pero no necesariamente más explicativos .…”

Population structure of Japanese Spanish mackerel Scomberomorus niphonius in the Bohai Sea, the Yellow Sea and the East China Sea: evidence from random forests based on otolith features

“…We selected RF instead of other available methods because of its: (1) easy usage and powerful ability that outperforms other classifiers; (2) no requirement of distributional assumption and independence of the predictor variables; (3) no overfitting [15].…”

“…However, few efforts have been put into the improvement of classification methods, whereas machine-learning methods (e.g. random forests, artificial neural networks) have been applied in fish parasites and otolith microchemistry, and have outperformed discriminant analysis [13][14][15]. Discriminant analysis (linear, quadratic and canonical) is commonly used in population studies based on otolith shape, and multivariate normality and independence of the predictor variables are required for this method.…”

An attempt to use ectoparasites as tags for habitat occupancy by small mammalian hosts in central Europe: effects of host gender, parasite taxon and season

“…Following Pérez-del-Olmo et al . (2010), each data set was randomly split into a training dataset (80% of observations) and an independent validation dataset (20% of observations). The configuration for building RF models included 500 trees and default number of a randomly selected set of predictor variables (square root of the total number of predictor variables).…”

“…2008; Pérez-del-Olmo et al . 2010). We believe that these studies are the only examples of the application of RF to parasitological data.…”