3.2- The influence of read sampling depth on taxon accumulation and α diversity
Read sampling depth strongly influences the number of taxa counted in an individual PCR replicate (observed alpha diversity). Taxon accumulation curves (Figure 1) show that for all PCR replicates, observed α diversity increases with read sampling depth across the range of depths used. On average across PCR replicates and sites, a read depth of 1,000 included 51% (PITS) and 33% (FITS) of the observed α diversity compared to that detected with 20,000 reads. A read depth of 5,000 retained 77% (PITS) and 64% (FITS) of taxa detected at 20,000 reads, and a read depth of 10,000 retained 89% (PITS) and 82% (FITS) of taxa detected with 20,000 reads. In all sites, the PITS curves surpass the inflection point where slope begins to decrease (asymptote) at a sampling depth under 5000 reads. In FITS, the inflection point is less visually observable and more taxa continue to accumulate at high read depth (Figure 1). Nonetheless, we observed that in both PITS and FITS datasets, the maximum observed richness values vary considerably across PCR replicates.
Because the maximum observed richness could be influenced by read sampling depth, we explored PCR replicate richness variation using asymptote values extrapolated to twice the original depth withiNEXT . We found Observed diversity estimates were rarely normally distributed and variance was high, with up to five replicates per group being outliers from the mean (Table S3). After outlier removal, PCR replicate richness at the extrapolated asymptote still exhibited multiple fold differences in PITS and standard deviations equivalent to up to 30% of the maximum richness of the group (Table 1). We found the highest fold differences in observed richness in the PITS data set from YL.1 (Figure 1e), a site situated within a marine lagoon at a location that is regularly inundated with both marine water and stream runoff. We expect these physical processes increase true taxon richness and possibly heterogeneity within the environmental samples. We observed fewer outliers in extrapolated richness for FITS, with only up to two outliers per group, but that variation was high, with standard deviations up to 21% of the maximum richness of the group. We observed the highest fold differences in observed richness in FITS at YL.1 and FO.2.
Increasing the read sampling depth from 1,000 to 10,000 reads resulted in an average 1.8-fold increase in observed α diversity for PITS and 2.4-fold increase for FITS (Table S4). This increase in observed α diversity with depth was significant for all six sites and both markers (Table S4). Shannon diversity did not significantly increase with read sampling depth for five of the six sites in the PITS data set, but did significantly increase with all FITS data set (Table S4). Simpson diversity did not significantly increase with read sampling depth for any PITS data set, but did for five of the six sites in the FITS data set (Table S4).