The terrestrial subsurface is a critical biome for the Earth’s biogeochemical cycling and sources of consumable freshwater, however, we still know little about the microbial communities that drive the biogeochemical processes and how they are impacted by environmental conditions (1). Environmental sequencing has provided the best avenue for studying these microbial communities, especially since many species have unknown conditions for lab culturing (2). In this study, we generated metagenomes to study the subsurface microbiome from the centimeter scale to the field scale (3) and to study the relationship between the sediment and water communities at the Oak Ridge Reservation, a site that has been impacted by heavy metal, nitrate, and radionuclide contamination due to leaching from waste disposal ponds (4, 5). We generated shotgun metagenomics data from 77 sediment samples and 33 groundwater samples. Metagenomes from the same sample were co-assembled, resulting in 26 sediment metagenomes and 9 groundwater metagenomes.
Sediment samples were taken from the boreholes EB106 and EB271, whose drilling, geology, location, and biogeochemistry are described in (6). Each borehole was cut into approximately 22.86 cm cores. From each core we took 5g replicates for DNA extraction using a modified protocol of the Qiagen PowerMax Soil kit as described in (7). For EB271, cores had 1–3 replicates. For EB106, only for 12 cores were we able to obtain enough DNA for metagenomics sequencing. The first 5 cores had 3 replicates. The other cores only have one replicate either due to poor DNA yield or failed metagenomics library prep, and some required ~ 20g of sediment to obtain enough DNA for sequencing, as indicated in Supplementary Table 1.
Water samples were collected from adjacent groundwater wells (FW106, FW115-24, GW271) to the boreholes one week after drilling. Well construction, location, and biogeochemistry are described in (6). On the day of sampling, 5 L of water was filtered sequentially in an anaerobic chamber through 0.2um and 0.1um PES filters. Water was also shipped to Lawrence Berkeley Lab overnight and biomass from 2 L of water was collected on 0.2um PES filters. DNA was extracted from filters using the same method as the sediment.
Metagenomics sequencing libraries were prepared using the Illumina Nextera Flex Kit (now called Illumina DNA Prep Kit), according to the manufacturer’s instructions. The library concentration was measured using a Qubit 3 Fluorometer. The samples were sequenced by the QB3 Genomics, UC Berkeley, Berkeley, CA (RRID:SCR_022170) using 2 × 150 bp on an Illumina HiSeq4000.
Illumina reads were filtered for phiX174 and trimmed to remove any residual adapter sequences using BBTools Version 38.86 (8) using parameters "bf1 ktrim = r k = 23 mink = 11” along with a database of Illumina adapters and phiX174 spike-in. The resulting set of reads was assembled using SPAdes Version 3.15.4 (9, 10) with parameters “—memory 1024 —only-assembler —meta -k 21,33,55,77,99,127”. The SPAdes built-in error correction was disabled as it isn’t generally necessary and required more computer memory than we had available.