Additional file 7: of The gut of the finch: uniqueness of the gut microbiome of the Galápagos vampire finch

Table S4. Description of dataset subdivisions for analysis of individual factors without confounding factors, taken by randomly sub-setting a sub-category of the dataset subdivision where appropriate, as indicated. The influence of season was analyzed using data subdivisions 1 and 3-8, thereby controlling for different species/island/diet groups and the singularity of the vampire finches. Diet was assessed with dataset subdivisions 9-12 by comparing all finch species sampled with the exception of the vampire finches binned into diet categories of herbivorous (including seeds, cactus, and plant matter) and omnivorous (including insects), with control for season and island embedded in the various subsets. Seed-eaters are not generally considered herbivores, thus these groupings were designed broadly to encompass plant matter versus other dietary foodstuffs. The uniqueness of sanguivory was tested in dataset subdivisions 13-15, where season, sample size, and phylogeny (through nearest relatives) were considered as well as a comparison with other finches whose feather isotopes were measured. The role of finch species grouping was analyzed in dataset subdivisions 3, 7, 12, 16, 18, and 19 to remove the confounding influence of the extreme diet of G. septentrionalis and to extract the largest groupings while avoiding potential biogeographical (island) and season bias and to assess the ground finches alone. The influence of island was assessed in dataset subdivisions 1-3, 6, 7, 12, and 17-19, with Santa Cruz, Santa Fé, and North Seymour sometimes grouped together as “Santa Cruz +” since these islands are in close geographic proximity (Fig. 1). In some cases (dataset subdivisions 1, 2, and 12), latitude was assessed, as the northern, more isolated islands of Pinta, Genovesa, Wolf, and Darwin versus the remaining islands of the Galápagos. (DOCX 135 kb