Improved Rumen Microbiome Interpretation via Empirical Filtration Derived from Known Mock Microbiomes

ABSTRACT

The rumen microbiome plays a critical role in ruminant health and performance. As a consequence, many studies investigate microbial community composition using 16S rDNA based sequencing. As such, data quality and data filtering are critical to accurately identify microbial community composition. In this study we use mock communities to empirically select data filtering parameters to reduce artifact populations and compared the effect of filtering using different bioinformatic pipelines using rumen bacterial community data. The filtering parameters identified provides consistent estimates of rumen microbial diversity, regardless of bioinformatic pipeline utilized and provides a more accurate view of microbiome structure and composition.