Test of two membrane filter types for biomass collectionin community profiling by metabarcoding

Abstract

With the rapid development of microorganism biodiversity estimation and community profiling via DNA metabarcoding, there is a growing need to identify key stages affecting the results of experiment. In order to ensure high-quality reproducible results, all potential sources of bias should be explored. While bioinformatic pipelines can be often corrected and optimized to recalculate results, the initial steps of sample collection procedure, potentially influencing the downstream results of bacterial community profiling, are much harder to correct easily, without repeating the whole experiment. We examined the differences of the amplicon bacterial community profiles generated from aquatic microbial communities captured by polycarbonate (PC) and cellulose acetate (CA) filters. The microbial community profiles were shown to be significantly different by biodiversity. When using a PC-filters, alpha- and beta-diversity of communities were significantly higher. Differential abundance analysis showed enrichment of major cyanobacterial OTUs in CA-treated communities. Based on these results, the use of PC-filters is the most optimal for bacterial community profiling by metabarcoding of the 16S rRNA gene fragments.