Intact vs. homogenized subsampling: testing impacts of pre-extraction processing of multi-species samples on invasive species detection

Abstract

With the increasing use of metagenetics for invasive species monitoring in aquatic habitats, fully assessing sample pre-processing is essential for appropriate data interpretation. This includes the impact of subsampling before DNA extraction, a common practice for processing zooplankton and other mixed samples where the volume of the sample exceeds the maximum allowed in a commercial DNA extraction kit. Our goal for this study was to assess the impact of these preprocessing methods on detecting invasive species through 1) examining operational taxonomic unit (OTU) richness assessments and 2) detecting low abundance OTUs in mixed samples. We used tagged amplicon high throughput sequencing (HTS) to amplify the mitochondrial cytochrome oxidase I (COI) gene from zooplankton samples collected from multiple bays. We then used a step-wise approach to assess the impacts of replication and homogenizing DNA extracts on the resulting OTU richness. We compared the results from 1) a single, intact subsample, 2) multiple, intact subsample replicates, 3) a single, homogenized subsample, or 4) multiple, homogenized subsample replicates. We expected OTU richness to be highest for multiple subsamples and for those homogenized, with subsequent impacts on rare OTU detection. Our results showed that homogenizing a sample prior to DNA extraction increased the number of OTUs recovered, particularly low abundance OTUs. Additionally, processing a greater number of extraction replicates increased the OTU richness of each sample, regardless of the extraction pre-processing conducted. However, with the increasing OTUs, more sequences per replicate were required to detect all OTUs. Thus, when detection of low abundance OTUs is a goal, particularly for early detection of invasive species, homogenizing multispecies samples is recommended due to the clear increase in rare OTUs detected.