Skip to main content
. 2019 Apr 29;19(4):863–876. doi: 10.1111/1755-0998.13012

Table 4.

Summary statistics of GAMM models relating Jaccard's distance and percentage matching between the composition of EPTO communities inferred from morphology and metabarcoding, in relation to subsampling day and DNA extraction methods

Jaccard's Distance Family Species
Parametric coefficients Estimate SE z‐value p Estimate SE t‐value p
Intersect 0.610 0.058 10.42 <2 × 10−16 *** 0.731 0.052 14.123 <2 × 10−16 ***
TISSUE 0.034 0.021 1.596 0.111ns 0.031 0.019 1.622 0.106ns
SOIL 0.067 0.022 2.987 0.003** 0.060 0.020 2.992 0.003**
Smooth terms edf   F‐value p edf   F‐value p
s(day): BEAD 1.000   7.411 0.007** 1   13.550 2.6 × 10−4 ***
s(day): TISSUE 2.205   21.612 5.5 × 10−10 *** 2.455   26.190 3.3 × 10−12 ***
s(day): SOIL 2.032   16.302 1.2 × 10−7 *** 1.381   31.240 1.5 × 10−8 ***
Percentage Matching Family Species
Parametric coefficients Estimate SE z‐value p Estimate SE t‐value p
Intersect 0.520 0.068 7.685 1.2 × 10−13 *** 0.303 0.057 5.277 2.1 × 10−7 ***
TISSUE −0.052 0.032 −1.634 0.103 ns −0.048 0.026 −1.839 0.067 ns
SOIL −0.098 0.033 −2.997 0.003** −0.094 0.027 −3.514 4.9 × 10−4 ***
Smooth terms edf   F‐value p edf   F‐value p
s(day): BEAD 1   5.565 0.019* 1   17.270 3.9 × 10−5 ***
s(day): TISSUE 2.315   26.032 2.9 × 10−12 *** 2.458   34.130 7.7 × 10−16 ***
s(day): SOIL 2.173   23.066 1.2 × 10−10 *** 1.822   28.600 4.0 × 10−9 ***

Extractions were performed from the ethanol used to preserve five unprocessed freshwater macroinvertebrate bulk samples and subsampled on days 1, 2, 3, 5, 7 and 14 after field sampling, using three DNA extraction methods (BEAD, TISSUE and SOIL). For each model, we provide the parameter estimates, standard errors (SE) and statistical significance of parametric terms, and the effective degrees of freedom (edf) and approximate significance of smooth terms. The shape of the smooth terms is provided in Figure 4. The matrix used to build the models excluded rare species with a percentage read count <0.01% from each sample; models build using alternative criteria for dealing with rare species are given in Tables S7 and S8.

***

p < 0.001.

**

p < 0.01.

*

p < 0.05.

ns

> 0.05.