Food from faeces: Evaluating the efficacy of scat DNA metabarcoding in dietary analyses

Abstract

Scat DNA metabarcoding is increasingly being used to track the feeding ecology of elusive wildlife species. This approach has greatly increased the resolution and detection success of prey items contained in scats when compared with other classical methods. However, there have been few studies that have systematically tested the applicability and reliability of this approach to study the diet of large felids species in the wild. Here we assessed the effectiveness of this approach in the cheetah Acinonyx jubatus. We tested how scat degradation, meal size, prey species consumed and feeding day (the day a particular prey was consumed) influenced prey DNA detection success in captive cheetahs. We demonstrated that it is possible to obtain diet information from 60-day old scats using genetic approaches, but the efficiency decreased over time. Probability of species-identification was highest for food items consumed one day prior to scat collection and the probability of being able to identify the species consumed increased with the proportion of the prey consumed. Detection success varied among prey species but not by individual cheetah. Identification of prey species using DNA detection methods from a single consumption event worked for samples collected between 8 and 72 hours post-feeding. Our approach confirms the utility of genetic approaches to identify prey species in scats and highlight the need to account for the systematic bias in results to control for possible scat degradation, feeding day, meal size and prey species consumed especially in the wild-collected scats.

A trophic latitudinal gradient revealed in anchovy and sardine from the Western Mediterranean Sea using a multi-proxy approach

“…It should be noted that DNA metabarcoding have some limitations when is used in diet-related studies, such as DNA degradation from the time of feeding until the sample collection, DNA extraction efficiency, selection of molecular markers, PCR inhibition or uncompleted reference databases 72 – 74 . In addition, DNA metabarcoding without microscope analysis does not determine the trophic position of each detected prey species in the food web.…”

eDNA metabarcoding for biodiversity assessment, generalist predators as sampling assistants

“…While this aspect has to be accounted for when interpreting eDNA metabarcoding results for diet assessments, it has useful implications in the context of biodiversity surveys. However, taxa detectability varies in metabarcoding approaches as well, due to a number of widely acknowledged biological and technical reasons 45 , 62 that should be considered when designing a study and drawing inferences.…”

“…For this analysis, we used the function prop.test , which is a 2-sample test for equality of the proportion of sequences identified to each taxonomic level (order, family, genus and species) between the two sample age categories. The two age categories are meant to serve as a broad separation between relatively “fresh” and “older” samples that are likely to be significantly affected by DNA degradation as a result of the cumulative effect of multiple factors acting during sample exposure to the environment (see also 36 , 45 ).…”

DNA metabarcoding successfully quantifies relative abundances of arthropod taxa in songbird diets: a validation study using camera-recorded diets

“…Validation studies using mock communities or captive animals fed a known diet, have been conducted for a wide range of consumers and prey and have shown that although broad correlations are likely, especially at level of the presence/absence of prey taxa, biases may occur that need to be accounted for (reviewed in Deagle et al (2019), see also (Thuo et al, 2019). Therefore Deagle et al (2019) recommended to incorporate cross-validation in a study setup whenever possible.…”