Dropping Hints: Estimating the diets of livestock in rangelands using DNA metabarcoding of faeces

Abstract

The introduction of domesticated animals into new environments can lead to considerable ecological disruption, and it can be difficult to predict their impact on the new ecosystem. In this study, we use faecal metabarcoding to characterize the diets of three ruminant taxa in the rangelands of south-western New South Wales, Australia. Our study organisms included goats (Capra aegagrus hircus) and two breeds of sheep (Ovis aries): Merinos, which have been present in Australia for over two hundred years, and Dorpers, which were introduced in the 1990s. We used High-Throughput Sequencing methods to sequence the rbcL and ITS2 genes of plants in the faecal samples, and identified the samples using the GenBank and BOLD online databases, as well as a reference collection of sequences from plants collected in the study area. We found that the diets of all three taxa were dominated by the family Malvaceae, and that the Dorper diet was more diverse than the Merino diet at both the family and the species level. We conclude that Dorpers, like Merinos, are potentially a threat to some vulnerable species in the rangelands of New South Wales.

DNA traces the origin of honey by identifying plants, bacteria and fungi

“…To remove possible misassigned reads and false positives, due to tag jumping or contamination, we followed a conservative approach of further filtering for all reads and OTUs (following e.g. 66 , 67 ). This approach included three means, as follows.…”

Estimation of the relative abundance of species in artificial mixtures of insects using low-coverage shotgun metagenomics

“…For instance, MEGAN (Huson et al 2007) and many other studies always ignore singletons. Other studies increase the minimum number of reads to keep a taxon in the list (five in Piñol et al 2014;ten in Gibson et al 2015 andin Lee et al 2018). As we do here, Pompanon et al (2012) and Alberdi et al (2017) suggest that a relative threshold can be more appropriate than absolute read count thresholds.…”

A multi-kingdom metabarcoding study on cattle grazing Alpine pastures discloses intra-seasonal shifts in plant selection and faecal microbiota

“…This last method proved to be one of the most accurate and versatile for characterizing diet composition 27 , especially when the diet cannot be defined morphologically 22 . It relies on the sequencing of target DNA fragments from the residues (mainly in the animal faeces) and on the matching to a database of known sequences to identify the taxonomic origin 12 , 28 , 29 . This technique, thanks to the advent of next generation sequencing platforms, offers fast processing and turnaround times, with a very low risk of subjective interpretation.…”

“…This technique, thanks to the advent of next generation sequencing platforms, offers fast processing and turnaround times, with a very low risk of subjective interpretation. Thus, DNA-based studies have examined the botanical composition of diets selected by domestic and wild ruminants, as for example sheep ( Ovis aries ) 28 , 29 , goat ( Capra hircus ) 30 , white-tailed deer ( Odocoileus virginianus ) 9 , and American bison ( Bos bison ) 31 , or the dietary niche overlap between different species 12 , 30 , 32 . Additionally, the main advantage that makes this technique very attractive is that it allows investigating whatever mix of biological entities, including plant, bacterial and fungal taxa.…”

“…In fact, using kingdom-specific primer pairs, it is virtually possible to recognize all the taxa composing a complex matrix. The rbcL , matK , trnL loci, alone or in combination, are largely applied for plants 27 , 29 , 33 , while the bacteria kingdom is almost exclusively investigated through the 16S rRNA region sequencing 34 , and the Internal Transcribed Spacer (ITS) is commonly used for fungal identification 35 .…”