This data package includes the abundance of microbial operational taxonomic units (OTUs) for samples collected during the austral summer of 2012-2013 in the Lake Hoare and Goldman Glacier Basins of Taylor Valley, Antarctica. A total of twenty samples from on- and off-water track soils were collected and analyzed. Samples were collected from the Lake Hoare Basin on 27 December 2012 and from the Goldman Glacier Basin on 4 January 2013. The aim of the study was to identify how variation in the measured physical and chemical environment of water tracks within the two water track systems influenced soil microbial community structure and diversity. Soil bacterial biodiversity was assessed using cultivation independent 16S rRNA gene sequencing.
...or query the dataset by variables and dates of interest:
As described in the associated manuscript, soil samples and pore water were collected from the upper 10 cm of the soil horizon using aseptic techniques, and were stored in sterile Whirl-Pack bags at -20°C until processing. Sediment and pore water collected from the darkened portions of water tracks were designated “on-track,” and samples from adjacent lighter soils were classified as “off-track.” Off-track samples were located at least 5 m from the current edge of the water tracks. Wet, on-track soils have a typical albedo of 0.15, while off-track soil albedo is generally 0.22, making them readily distinguishable in the field.
DNA extraction and microbial community analyses were conducted using the cultivation-independent 16S rRNA gene sequencing approach as described in Prober et al., 2015 (doi: 10.1111/ele.12381). Total genomic DNA was extracted from each sample using the MoBio PowerSoil DNA Isolation Kit. For microbial analyses, the 4v hypervariable region of the 16S rRNA gene was PCR amplified using the 515f and 806f primer pair which captures both Bacteria and Archaea. Three PCRs were run per sample, with the amplicons from the replicate reactions pooled. Each primer pair included Illumina adapters and 12-bp error-correcting barcodes unique to each sample, as described in the Earth Microbiome Project protocol (Thompson et al., 2017; doi: 10.1038/nature24621). After gel visualization to confirm amplification, PicoGreen dsDNA assay was used to quantify amplicon yields. The amplicons were then pooled together in equimolar concentrations for sequencing on the Illumina MiSeq instrument. DNA sequencing was completed at the University of Colorado Next Generation Sequencing Facility using the 2x150pb paired-end sequencing chemistry. Four DNA extraction and for no-template PCR ‘blanks’ were included in the run to check for potential contamination.
Sequences were demultiplexed using a custom Python script (‘prep_fastq_for_uparse.py’, at: https://github.com/leffj/helper-code-for-uparse), with the UPARSE pipeline used for quality filtering and phylotype (i.e. operational taxonomic unit) clustering. Quality filtering was conducted using a maximum e-value of 0.5 with paired-end sequences merged prior to downstream processing. Representative sequences from returned phylotypes that were not ≥75% similar to sequences contained in the Greengenes database were removed; afterwards the raw sequences were mapped to phylotypes at a 97% similarity cutoff. Taxonomic classification of each phylotype was determined using the Ribosomal Database Project classifier against the Greengenes database with a confidence threshold of 0.5. The OTU table for which this metadata describes is the result.
The OTU table and eDNA sequences used for this study were from a legacy project, from which only the OTU table remained.
Funding for this study was provided by the National Science Foundation (NSF) as follows: