Results:
We detected Bd DNA on all Sierra Nevada yellow-legged frogs (n=6)
swabbed for comparison testing across all three sites using the field
and lab protocols (Table 1 & Fig. 3). Of the anurans with unknown Bd
levels, we detected Bd DNA on swabs from 75% of the R. draytonii(n=4) as well as the L. catesbeianu s (n=1) using the field
protocol. We did not detect Bd on one R. draytonii swab. Lab
analysis of the R. draytonii and L. catesbeianus swabs verified
the negative and positive results.
We detected Bd DNA in eDNA samples collected at two out of the three
sites using both the lab and the field protocols. We did not detect Bd
DNA for Site 3 using the field protocol and we did not detect Bd DNA for
Site 2 using the lab protocol (Table 1 & Fig. 3).
We were unable to quantify DNA from both the eDNA and swab samples using
the field protocol because all DNA levels were below the standard curve
(<100 copies). All positive Bd eDNA and swab samples extracted
and analyzed using the lab protocol were >100 copies and
within the limits of quantification (Table 1).
We detected Bd in more technical replicates using the lab protocol
compared to the field protocol (Fig. 3 and Fig. 4). The best fit
detection probability model with the lowest PPLC and WAIC included
sample type (swab vs eDNA) and analysis approach (field vs. lab) as
covariates for the sample (θ) and sub-sample (p ) respectively.
The mean derived estimated of θ was higher for swab samples compared
eDNA samples, as expected, and the mean derived estimated of pwas higher for lab protocol compared to the field protocol (Fig. 4). The
mean θ (± 95% credible limits) was 0.47 (0.42-0.59) for the eDNA
samples and 0.80 (0.70-0.94) for skin swabs. The mean p (± 95% credible
limits) was 0.65 (0.49-0.75) for the field protocol and 0.97 (0.90-0.99)
for the lab protocol. Approximately 1.5 technical replicates would have
to be analyzed using the field protocol in order to have the same mean
detection probability of one technical replicate using the lab protocol.
Reduced detection probability of the field-based approach is compounded
when collected as eDNA samples, where 1.7 technical replicates would
need to be analyzed in order to have the same mean detection probability
of one technical replicate from a frog swab.
While using the field protocol, one skin swab and one eDNA sample
experienced inhibition at Site 3. To resolve, we removed the inhibition
in the skin swab and eDNA sample through two serial dilutions (1:1) with
molecular grade water (final dilution: 10 µl DNA and 30 µl
H20). We detected Bd DNA in the skin swab but failed to
detect Bd DNA in the eDNA sample. However, we did not detect Bd DNA in
any of the uninhibited eDNA samples collected at Site 3 using field
analysis techniques. There was no issue with inhibition for all samples
extracted and analyzed using the lab protocol.
All the negative control samples, extraction negatives, and qPCR
negatives tested negative for both the lab and field protocols.