Uncovering Effects from the Structure of Metabarcode Sequences for Metagenetic and Microbiome Analysis

Abstract

The advent of next-generation sequencing has allowed for higher-throughput determination of which species live within a specific location. Here we establish that three analysis methods for estimating diversity within samples—namely, Operational Taxonomic Units; the newer Amplicon Sequence Variants; and a method commonly found in sequence analysis, minhash—are affected by various properties of these sequence data. Using simulations we show that the presence of Single Nucleotide Polymorphisms and the depth of coverage from each species affect the correlations between these approaches. Through this analysis, we provide insights which would affect the decisions on the application of each method. Specifically, the presence of sequence read errors and variability in sequence read coverage deferentially affects these processing methods.

Life barcoded by DNA barcodes

“…These biological entities next can be compared with OTUs or ASVs in different studies, such as the BIN framework introduced by BOLD, to estimate the biodiversity of target samples. Yet biological interpretation of metabarcoding data can be seriously affected by the differences between the two methods: OTUs minimize the effects of slight variations in sequences that may or may not be of interest, but a small change, as in the case of parasitoid wasps, could be capturing actual differences between species; on the contrary, ASVs are defined as all “unique reads” within a metabarcoded dataset, often leading to a wrong differentiation between the SNPs of the same species, and in the same way making sequencing or PCR errors more prominent when compared to OTUs (Molik et al 2020 ). By using simulations, it has been advised that approaches utilizing ASVs outperform OTUs only when the sequencing depth is sufficient to cover a biological complexity with low polymorphisms.…”

Systematic Comparisons for Composition Profiles, Taxonomic Levels, and Machine Learning Methods for Microbiome-Based Disease Prediction

“…ASVs are commonly generated using the Divisive Amplicon Denoising Algorithm 2 (DADA2), and the resultant ASVs represent true biological sequences obtained from reads (Callahan et al, 2016). In addition, there have been recent efforts to use the occurrence of short-chain k-mer (15-30mer) (Molik et al, 2020), and very short-chain k-mers (<10mer) (Asgari et al, 2018(Asgari et al, , 2019, within reads that offer a unique reference-free and alignment-free approach to provide a data representation upon which a phenotype prediction model is built. We have included both of these k-mer approaches in our review to compare them directly with the OTU/ASV assignment methods.…”

Using metatranscriptomics to estimate the diversity and composition of zooplankton communities

“…For mock community analysis, metatranscriptomics provided the most reliable species diversity and community composition estimates, which closely resembled those derived from morphological data. The use of metatranscriptomics avoided the co‐detection of extra‐organismal eDNA and minimized background noise encountered during PCR‐based methods, which may cause inflated estimates of species richness and complicated taxonomic assignment of sequences, especially with the absence of good quality reference databases (Molik et al, 2020 ). Another advantage of using RNA for monitoring zooplankton is that it avoids the bias related to NUMT pseudogene contamination (Collura et al, 1996 ).…”