Fast model-based ordination with copulas

Popović, Gordana; Hui, Francis K. C.; Warton, David I.

doi:10.1101/2021.03.28.437086

Cited by 3 publications

(2 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This process has been implemented within the ecopower package, using an internal function called extend. This function takes a cord object (obtained by fitting a Gaussian copula to a manyglm object using the cord function from the ecocopula package; Popovic et al, 2021) and simulates N multivariate abundances using the above procedure (to date, it can handle Poisson, negative binomial and binomial distributions). The function then refits the simulated responses to a manyglm object with a data frame that is 'extended' in a manner that preserves the original or prespecified design (Box 2).…”

Section: Challenge 1 -Data Generating Modelmentioning

confidence: 99%

How many sites? Methods to assist design decisions when collecting multivariate data in ecology

Maslen¹,

Popović²,

Vergés³

et al. 2022

Preprint

View full text Add to dashboard Cite

1. Sample size estimation through power analysis is a fundamental tool in planning an ecological study, yet there are currently no well-established procedures for when multivariate abundances are to be collected. A power analysis procedure would need to address three challenges: designing a parsimonious simulation model that captures key community data properties; measuring effect size in a realistic yet interpretable fashion; and ensuring computational feasibility when simulation is used both for power estimation and significance testing.2. Here we propose a power analysis procedure that addresses these three challenges by: using for simulation a Gaussian copula model with factor analytical structure, fitted to pilot data; assuming a common effect size across all taxa, but applied in different directions according to expert opinion (to "increaser", "decreaser" or "no effect" taxa); using a critical value approach to estimate power, which reduces computation time by a factor of 500 with little loss of accuracy.3. The procedure is demonstrated on pilot data from fish assemblages in a restoration study, where it was found that the planned study design would only be capable of detecting relatively large effects (change in abundance by a factor of 1.5 or more).4. The methods outlined in this paper are available in accompanying R software (the ecopower package), which allows researchers with pilot data to answer a wide range of design questions to assist them in planning their studies.

show abstract

Section: Challenge 1 -Data Generating Modelmentioning

confidence: 99%

How many sites? Methods to assist design decisions when collecting multivariate data in ecology

Maslen¹,

Popović²,

Vergés³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…An analysis of Deviance (Dev) was performed to test the fitness of the model, with 999 bootstraps iterations as a resampling method (Davison and Hinkley, 1997), using the function anova.manyglm from the package mvabund (Wang et al, 2017). Moreover, an ordination to visualize the differences in community composition between the two seasons (BH and CR) and the two sub-colonies or sites (NP and PL) was computed and plotted using the cord() function from the R package ecoCopula (Popovic et al, 2021).…”

Section: Biodiversity Analysesmentioning

confidence: 99%

Metabarcoding faecal samples to investigate spatiotemporal variation in the diet of the endangered Westland petrel (Procellaria westlandica)

Querejeta

Lefort

Bretagnolle

et al. 2020

Preprint

View full text Add to dashboard Cite

As top predators, seabirds are directly impacted by any changes in marine communities, whether they are linked to climate change or caused by commercial fishing activities. However, their high mobility allows them to adapt to changing conditions. For example, seabirds can adapt their foraging behaviour according to the resources available at different seasons. This capacity of adaptation comes to light through the study of their diet. Traditionally, the diet of seabirds is assessed through the morphological identification of prey remains in regurgitates, a method that is invasive for the bird and limited in terms of resolution. However, the recent optimization of DNA-based approaches is now providing a non-invasive and more comprehensive and accurate characterization of animals' diet. Here, we used a non-invasive metabarcoding approach to characterize the diet of the Westland petrel (Procellaria westlandica), an endangered burrowing species, endemic to the South Island of New Zealand. We collected 99 fresh faecal samples at two different seasons and in two different sub-colonies. Besides from describing the diet of the Westland petrel, our aim was to account for seasonal and geographical variations in the diet of the petrel and assess potential links with the fishery industry in New Zealand. We found that amphipods were the most common prey, or secondary prey, followed by cephalopods and fish, suggesting a close link between the composition of prey items and New Zealand's commercial fishing activities but, also, some level of natural predation. Our results show significant differences in diet between seasons (before hatching vs chick rearing season) and between sampling sites (two sub-colonies 1.5 km apart), which suggests variability and adaptability in the foraging strategy of the Westland petrel. Due to its non-invasive nature, the method used here can be applied on a great number of samples to draw a comprehensive picture of the diet dynamic in seabirds and unravel their adaptability or ecological requirements. This work demonstrates how environmental DNA can inform the conservation of an endangered species with elusive foraging behaviour, providing, in this case, valuable information regarding the diet preferences of an iconic species within New Zealand's biodiversity. Introduction

show abstract

Fast model‐based ordination with copulas

2021

Self Cite

View full text Add to dashboard Cite

Visualisation is a key step when working with data in ecology, both for exploration, and to support conclusions from statistical analyses.Many studies in ecology however collect multivariate and discrete data, which are difficult to visualise. In this article, we focus on ordination (Legendre & Legendre, 2012;Zuur et al., 2007), a general and very commonly applied class of visualisation methods which aim to collapse or project multivariate data onto a small number (often two or three) dimensions in ways that preserve as much of the underlying structure as possible, such that it can be plotted to look for prevailing patterns. In this paper, our focus is on ordination methods for multivariate abundance data, for example species assemblages, although in principle the ideas discussed here would apply more generally to any multivariate analysis. There are two broad categories of ordination methods in the current literature, traditional dissimilarity-based methods and modelbased methods. Dissimilarity-based ordination methods begin by

show abstract

Fast model-based ordination with copulas

Cited by 3 publications

References 42 publications

How many sites? Methods to assist design decisions when collecting multivariate data in ecology

How many sites? Methods to assist design decisions when collecting multivariate data in ecology

Metabarcoding faecal samples to investigate spatiotemporal variation in the diet of the endangered Westland petrel (Procellaria westlandica)

Fast model‐based ordination with copulas

Contact Info

Product

Resources

About