Skip to main content
. 2021 Sep 30;9:e12065. doi: 10.7717/peerj.12065

Table 2. Comparison of taxonomy assignments of 16S and ITS USEARCH ASVs for urban parks bioaerosol communities with the RDP and SILVA (16S)/UNITE(ITS) taxonomy databases, and with a combined database approach.

Taxonomy database used RDP training set (16S) SILVA (16S) P-value
between 16S
databases
RDP training set (ITS) UNITE (ITS) P-value
Between ITS
databases
Metrics
Total genera identified by pipeline 517 694 0.02 501 696 0.06
Alpha diversity of genera (Hill numbers) Observed (D0) 112 127 0.30 121 123 0.75
expShannon (D1) 11.31 12.56 0.52 14 8.56 0.02
invSimpson (D2) 7.99 8.98 0.56 8.71 5.3 0.00
Beta diversity of genera (Bray–Curtis) Mean 0.78 0.8 0.56 0.84 0.82 0.61
Median 0.8 0.83 0.52 0.86 0.85 0.75
Range 0.08–1 0.09–1 NA 0.17–1 0.10–1 NA
Standard deviation 0.19 0.18 0.44 0.14 0.18 0.17
Unassigned ASVs at genus level Number 2,479 1,681 0.02 3,899 3,559 0.28
% total ASVs 56% 38% 0.00 71% 64% 0.01
Unassigned ASVs at genus level with both databases Number 1,011 2,871
% total ASVs 23% 53%

Note:

Observed (D0) is the raw diversity or number of unique ASVs detected. ExpShannon (D1) refers to the exponential of the Shannon diversity index and invSimpson (D2) refers to the inverse of the Simpson diversity index. Ranges are not appropriate for statistical testing so an N/A appears instead of a p value for them. P values that are significant at the 0.05 threshold appear in bold.