Poster Presentation Australian Microbial Ecology 2019

Accurately estimating a core community and using it as a tool to tease apart vertically and environmentally acquired components of the gut community of termites (#109)

Ghislaine Platell 1 , Katharine Howell 1 , Boris Baer 2 , Tamara Hartke 3
  1. ARC Centre of Excellence in Plant Energy Biology, University of Western Australia, Crawley, WA, Australia
  2. Centre for Integrative Bee Research, University of California, Riverside, California, USA
  3. Blumenbach Institute for Zoology and Anthropology, University of Göttingen, Göttingen, Germany

Our work on the variability of gut community structure and composition within a termite species has highlighted the need for extensive sampling and standardised techniques, to maximise data reproducibility and utility across research groups. Determining the core community has implications regarding taxa that may be vertically transferred from generation to generation and three aspects related to sampling for termite core microbiome studies were investigated: scale, the geographic distribution of sampling sites; scope, the caste (workers, soldiers, etc) composition of samples; and sampling intensity, the number of samples taken per unit of interest. Taxa shared by all castes are likely vertically transferred, whereas taxa exclusive to a location or feeding group are likely to be environmentally acquired, ingested with food or soil, particularly if these are shared across species. 

We found that sampling intensity had the largest effect on the outcome of core community calculations, and is fortunately the easiest factor for researchers to control. We developed a simple random sampling method which future researchers can use on their own data sets to test whether sufficient sampling intensity has been reached. Scale and scope also had an impact and we make the following sampling recommendations to obtain an "accurate" core community:

Scope: All relevant sample types (castes in the termite context) should be included, for example only workers if that is the scope of the question, or all castes for a total species core community estimation.

Scale: The maximum relevant range should be included, which could be a single colony if that is the scope of the question, or multiple locations across the entire species range for a total species core community estimation.

Intensity: A minimum of 20 (ideally 30) samples per level of each unit of interest should be included. This means 20 samples total to estimate a species core microbiome, or 20 samples per treatment, location or colony as relevant to the scope of the research question.