Data from: Are fecal samples an appropriate proxy for amphibian intestinal microbiota?

Bioinformatics work was performed using Quantitative Insights into Microbial Ecology 2 (QIIME 2, version 2022.11) (Bolyen et al., 2019) and R (R Core Team 2021). Raw sequence data were demultiplexed by Macrogen (Seoul, South Korea). We evaluated the quality of demultiplexed reads and trimmed to remove poor-quality bases. Sequence quality control and denoising were performed with DADA2 (Callahan et al., 2016) to generate amplicon sequence variants (ASVs) using q2-dada2. Representative sequences were aligned with MAFFT (Katoh and Standley, 2013) using q2-alignment. To maximize taxonomic classification accuracy, we assigned taxonomy to ASVs using a weighted Bayes classifier that incorporates information on environment-specific taxonomic abundance (Kaehler et al. 2019), pre-trained on the 16S 515F/806R region in SILVA 138 (Quast et al., 2013) (MD5: b9476399080d189b4c9917d1246e7c69) (Bokulich et al., 2018; Robeson II et al., 2021) using q2-feature-classifier (Bokulich et al., 2018). Sequences were removed if unassigned at the domain level or assigned to be mitochondria/chloroplast/archaea. The format of the metadata file used in QIIME2 was validated by the cloud-based Google Sheets add-on Keemei (Rideout et al., 2016). The output QIIME2 files were exported to R for subsequent analysis using R package “qiime2R” (Bisanz 2018).