Dataset: AT39-01 CathodicEET Amplicons
Deployment: AT39-01

The FastDNA kit from MP Biomedicals (Irvine, CA, USA) was used for all DNA extractions following manufacturers instructions with the exception of using a Retsch MM 400 shaker (frequency 20 r/S for 5 min) instead of vortexing as described in detail elsewhere (DAngelo, 2019). From the Echem and Fluid samples, the ITO swab end was cut into a sterile plastic tube and subjected to a freeze-thaw cycle (8 min at -80C then thawing at room temperature for 10 min or until ice barely melted into water) before transferring into a FastDNA kit tube for DNA extraction. For the Shipboard and NP12 samples, 0.232 g/l polyadenine (polyA, final concentration) was added to the sample at the first step to increase DNA yield from low biomass samples. For each batch of samples extracted, a no template control (NTC) was included as a check for contamination. DNA concentrations were measured by Q-bit Fluorometer 3.0 (Thermo Fisher Scientific, Waltham, MA USA) using the Qubit HS dsDNA kit (Thermo Fisher Scientific, Waltham, MA USA) according to manufacturer instructions. Aliquots of DNA extracts were sent to the Integrated Microbiome Resource facility at Dalhousie University (Halifax, Canada) for sequencing of the V4-V5 hypervariable region of the 16S rRNA gene via Illumina Mi-Seq technology using the 515F/926R primer pair. Sequencing was performed on an Illumina MiSeq using 300 300 bp paired-end V3 chemistry. Raw sequenced reads were processed at the same time as sequenced reads from mineral incubation experiments taken from the same environments, as well as additional no-template-control samples (data not shown, but full data available at the NCBI Sequence Read Archive BioProject PRJNA564565 samples SAMN12723399-489), using modifications of the standard DADA2 pipeline Version 1.12 to construct unique amplicon sequence variants (ASVs). This was done to directly link ASVs between the two datasets and to give the DADA2 algorithm the most possible data to use for sequence inference. Thirty base-pairs were trimmed off of the start of each read, forward reads were truncated at 250 base-pairs, and reverse reads were truncated at 200 base-pairs. This allowed for 50 BP of overlap for the forward and reverse reads to be merged after sequence inference. Approximately 100 million bases from ~500K reads were used by DADA2 to infer the base-call error rates in the data. Inference of ASVs, merging of forward and reverse reads, and chimera removal resulted in 2,011 ASVs ranging from 345 360 BP in length. These sequences were assigned taxonomy by the nave Bayesian classifier built into the DADA2 package using the Silva v132 database. Scripts used to process the data are available via github.