Comparison of microbial communities from diverse biological matrices using mock community as an in situ positive control

Metataxonomy has become the standard for characterizing the diversity and composition of microbial communities associated with multicellular organisms and their environment. Understanding the interactions between the microbiotas within the same ecosystem is essential for fully understanding the role of microorganisms in evolutionary and ecological processes; however, such comparative studies across diverse biological samples are rare. In particular, currently available protocols assume a uniform DNA extraction, amplification and sequencing efficiency for all sample types and taxa. The addition of a mock community (MC) to biological samples before the DNA extraction step could aid identification of technical biases, but the impact of MC on diversity estimates is unknown. Here, standardized aliquots of bovine faecal samples with high or low biomass were extracted with high or low doses of MC, characterized using standard Illumina technology for metataxonomics, and analysed with custom bioinformatic pipelines. We showed that a MC was an informative in situ positive control provided an estimate of 16S rRNA sample gene copies (which allowed a more direct measure of community size), and detected sample outliers. However, we also demonstrated that if the recommended dose of MC is added to a sample with low biomass, diversity estimates were distorted. Using our results, we recommend MC doses for a range of sample types, including rhizosphere soil, whole invertebrates, and vertebrate faecal samples. ImportanceThe simultaneous processing of the sample microbiota with a known number of readily identifiable MC cells (co-extracted with the sample cells) or SNA (co-amplified with the sample DNAs) can be a valuable in situ positive control. However, guidelines regarding their application are very limited and do not consider the effect of these controls on sample diversity estimates. We demonstrate that a MC co-extracted with the study sample provides several advantages, such as highlighting bias in DNA extraction of gram positive, identifying potential sample outliers and inferring the number of 16S rRNA gene copies in the sample. However, the incorporation of a MC requires prior knowledge of sample biomass, as high ratios of MC to sample microbiota lead to biased sample diversity estimates. Practical advice on determining the appropriate MC dose for a wide range of sample types, including rhizosphere soil, whole invertebrates and mammalian faecal samples are provided.