Data from: Using spatial-temporal filtering and improved barcoding tools to improve the ecological relevance of pollen meta-barcoding

DNA metabarcoding has been successful for the rapid identification of
species in complex ecological assemblages, including identifying
interspecific interactions among species. However, advances in
metabarcoding within the plant kingdom have been hampered due to a lack of
universal gene regions that work across all taxa, which limit the
applications of eDNA and metagenomics in ecology more broadly. To
circumvent these limitations, we propose a holistic spatio-temporal
approach that combines multi-gene barcoding with existing plant occurrence
databases, species distribution models, and phenological analyses to
generate a shortened list of candidate species to increase metabarcoding
accuracy and computing efficiency. To validate the accuracy and efficiency
of our methodological framework, we compared the results of the DNA
metabarcoding from pollen loads of several species of wild bumble bees to
in-depth, long-term field observations of bee-plant interactions, along
with expert-led pollen identification. We show that DNA metabarcoding of
the plant species included in bumble bee pollen loads was most accurate
when combined with a candidate taxa list of plant species flowering in the
community when the bumble bees were foraging, which improved the accuracy
and taxonomic precision of 77.5% of samples. With the recent proliferation
of species occurrence and phenology data in tandem with advances in
computing and software development, we believe that spatio-temporal
filtering provides a simple approach for interpreting metagenomic studies
globally. Additionally, we demonstrate that the Angiosperms 353 probes
(developed for phylogenomics) offer significant promise for metagenomics
projects globally, including metabarcoding to reveal species interactions
within complex communities. Further, our approach demonstrates that
integrating DNA metabarcoding is most accurate and powerful when combined
with local ecological data.