Accuracy of microbial community diversity estimated by closed- and open-reference OTUs

Abstract

Next-generation sequencing of 16S ribosomal RNA is widely used to survey microbial communities. Sequences are typically assigned to Operational Taxonomic Units (OTUs). Closed- and open-reference OTU assignment matches reads to a reference database at 97% identity (closed), then clusters unmatched reads using a de novo method (open). Implementations of these methods in the QIIME package were tested on several mock community datasets with 20 strains using different sequencing technologies and primers. Richness (number of reported OTUs) was often greatly exaggerated, with hundreds or thousands of OTUs generated on Illumina datasets. Between-sample diversity was also found to be highly exaggerated in many cases, with weighted Jaccard distances between identical mock samples often close to one, indicating very low similarity. Non-overlapping hyper-variable regions in 70% of species were assigned to different OTUs. On mock communities with Illumina V4 reads, 56% to 88% of predicted genus names were false positives. Biological inferences obtained using these methods are therefore not reliable.

Keywords: Alpha diversity; Beta diversity; Closed-reference; OTU; Open-reference; QIIME.

Figure 1. Rarefaction curves for Bok reads generated by QIIME.

There are two Even and two Staggered samples of Mock3 (22 strains). The e parameter is the number of reads per sample.

Figure 2. Distribution of closed-reference beta diversities for all pairs of Mock2/3 samples.

The histograms show the distribution of weighted Jaccard (A, C) and weighted UniFrac (B, D) distances on all pairs of samples containing Mock2 or Mock3. A zero value for the Jaccard or UniFrac distance indicates maximum similarity between a pair of samples; one indicates maximum difference. Histograms (A) and (B) show the distribution when the same tag is sequenced (e.g., V4), histograms (C) and (D) when different tags are sequenced (e.g., V13 and V69). The y axis is the frequency, calculated as (number of sample pairs having distances which fall into a given bin) divided by (total number of sample pairs).