GoKawiilSample collection
Samples were collected under the permit G12/35236–1 issued by the Great Barrier Reef Marine Park Authority. Seawater samples were collected from 48 coral reefs spanning the length of the GBR, aligning with in situ coral surveys by the AIMS LTMP in 2019 and 2020 (Fig. 1). Four 5-l biological replicates of seawater were collected above the reef (2–10 m depth) at each site (total n = 192) in either Nalgene or Niskin bottles and placed on ice until return to the vessel. Each 5 l volume was independently filtered successively through 5-µm, 33 mm diameter cellulose acetate and 0.22-µm Sterivex-GP filters (Millipore). The Sterivex filters were then snap frozen and stored at −75 °C until DNA extraction.
In addition to metagenomics data, a total of 17 physicochemical variables were measured as part of this survey on 3 of the 4 replicates (5 l each) from each site, including ammonia (NH 4 +), nitrite (NO 2 −), nitrate (NO 3 −), total dissolved nitrogen, phosphate (PO 4 3−), total dissolved phosphorus, dissolved organic carbon, silicate, total suspended solids, chlorophyll a pigment concentration, phaeophytin A, particulate organic carbon, particulate nitrogen, particulate phosphorus, temperature, salinity and chlorophyll a fluorescence, using methods described in Terzin et al.76.
DNA extraction
DNA extractions were carried out inside each 0.22-µm Sterivex filter by adding 1.8 ml filter-sterilized lysis buffer containing 50 mM Tris-HCL (pH 8.0), 40 mM EDTA (pH 8.0), 256 mg ml−1 sucrose and 18 µl lysozyme (100 mg ml−1). The filters were incubated with mild agitation for 1 h at 37 °C, followed by the addition of 20 µl proteinase K (20 mg ml−1) and further incubation and agitation for 1 h at 55 °C. The lysate was then ejected through the Sterivex into a 5 ml microtube and an equal volume of phenol:chloroform:isoamyl alcohol (IAA; 25:24:1) was added, mixed by inversion and centrifuged at 16,000g for 10 min. The aqueous phase was then recovered and an equal volume of chloroform:IAA (24:1) was added and mixed by inversion, then centrifuged for 10 min at 16,000g. DNA was precipitated by adding 1 ml isopropanol and recovered by centrifugation at 20,000g for 25 min and the supernatant discarded. The DNA was washed with 500 µl 70% ethanol, centrifuged for 10 min, the ethanol discarded and the pellet air-dried. Finally, 20 µl PCR water was added, incubated at 4 °C overnight to resuspend the DNA pellet and frozen at −20 °C.
Illumina sequencing
All 191 replicates from the 48 sites were subjected to Illumina sequencing at the Centre for Microbiome Research (Brisbane, Australia). Libraries were prepared using the Nextera DNA Flex Library Preparation Kit (Illumina 20018705) according to manufacturer’s protocol. Libraries were pooled at equimolar amounts of 2 nM per library. The library pool was quantified in triplicates using a Qubit dsDNA HS Assay Kit (Invitrogen) and sequenced using an S4 2× 150 bp flow cell on a NovaSeq 6000. Additional ‘deep’ sequencing was carried out for one replicate from 27 of the 48 sites to a target depth of 40 Gbp for Nanopore polishing and hybrid assembly (Supplementary Tables 1 and 2).
Nanopore sequencing, basecalling and quality control
The replicate chosen for Illumina deep sequencing was also submitted for long-read sequencing on Oxford Nanopore Technology’s (ONT) PromethION platform at the Australian Centre for Ecogenomics (Brisbane, Australia) sequencing facility. For ONT sequencing, DNA was first size-selected using the Circulomics SRE XS kit (PacBio, SKU 102-208-200) as this was shown in a small trial to increase total data output and read N50. Native barcoded libraries were prepared from the size-selected DNA following a genomic DNA by ligation protocol (Oxford Nanopore, LSK-109 with EXP-NBD104). Barcoded libraries were pooled with two samples per PromethION flow cell at equal concentration and the library pool was sequenced on a PromethION 24 (Oxford Nanopore) for a total of 72 h with a R9.4 flow cell using MinKNOW (v20.06.18) with default settings. We then performed rebasecalling of the fast5 files using the superaccuracy model with Guppy (v5.0.16) to generate higher accuracy raw reads before assembly. Finally, Porechop (https://github.com/rrwick/Porechop; v0.2.4) was used for barcode trimming using default settings.
Hybrid metagenome assembly
... continue reading