Pelagic metagenome-assembled genomes from an estuarine salinity gradient in San Francisco Bay

Despite the ecological, cultural, and economic importance of San Francisco Bay (SFB), few studies have published microbial metagenome-assembled genomes (MAGs) from SFB (1, 2). We submitted eight bottom water samples from along the salinity gradient for metagenomic sequencing particularly to gain insights into nitrifier ecology in ammonia-rich SFB waters. Using a variety of methods, we created a library of representative MAGs from riverine to brackish to marine waters of SFB. We previously reported on ammonia-oxidizing archaea MAGs, including a massive bloom lineage captured by this data set (1), and the ecogenomics of Pelagibacterales (SAR11) subclade IIIa (2); however, most of the rich genomic data we generated remain to be explored.

Sample collection, DNA extraction, sequencing, and assembly methods were previously described (1). Briefly, 400–1,000 mL of bottom water (1 m above estuary floor) was collected from the SFB channel on 24th and 25th October 2013, at eight stations during United States Geological Survey (USGS) Water Quality monitoring cruises, including stations 657 (fresh, <0.5 PSU), 649 (oligohaline, 0.5–5 PSU), 3 (mesohaline, 5–18 PSU), 9 and 13 (polyhaline, 18–30 PSU), and 18, 27, and 34 (euhaline, 30–40 PSU). Microbial biomass was collected on 0.22-µm filters after passage through a 10-µm pore size pre-filter, and DNA was extracted using the FastDNA SPIN Kit for Soil (MP Biomedicals, Santa Ana, CA). Library preparation and metagenomic sequencing were performed by the DOE Joint Genome Institute (JGI) via a CSP project (Proposal ID 503022) to create Illumina fragments (300 bp) sequenced on an Illumina HiSeq 2500-1TB generating paired 150 bp length reads. For library preparation, 100 ng of DNA was sheared (to 479–610 bp) using the Covaris LE220 and size was selected using SPRI beads (Beckman Coulter). The fragments were treated with end-repair, A-tailing, and ligation of Illumina-compatible adapters (IDT, Inc.) using the KAPA-Illumina library creation kit (Kapa Biosystems). qPCR was used to determine the concentration of the libraries. We used quality-controlled and filtered metagenome data generated by JGI using their standard BBtools (v38.87) (3) pipeline (v3.7.3) for assembly, binning, and refining using the metaWRAP (v1.3.2) pipeline (4). Additionally, metagenomes were subset using Seqtk (https://github.com/lh3/seqtk) to 5%, 10%, 20%, and 50% of reads and co-assembled based on salinity zone. All metagenomes, subsets, and co-assemblies were assembled and binned as previously described (1) using metaSPAdes (v3.13.0) (5), MEGAHIT (v1.1.3) (6), MaxBin 2.0 (v2.2.6) (7), and MetaBAT2 (v2.12.1) (8). Bins were consolidated and filtered using metawrap bin_refinement to have >50% completeness and <10% contamination and then reassembled with metaSPAdes. MAGs were dereplicated using dRep (v2.3.2) (9) at 95% average nucleotide identity (ANI) and then taxonomically classified using the Genome Taxonomy Database toolkit (GTDB-tk v2.1.1; GTDB R07-RS207) (10).

Here, we present a total of 449 MAGs representing 17 phyla, including many MAGs from Proteobacteria (n = 178), Bacteroidota (n = 118), and Actinobacteriota (n = 73) (Fig. 1). Binning from subset metagenomes yielded MAGs from high-coverage genera such as Pelagibacter, Nitrosopumilus, or TMED189 (Actinobacteria) (Fig. 2). A median of 10.6× (range 4× to 103×) genome coverage was optimal for recovering genomes. An average of 24% (range 19%–28%) of metagenomic reads were competitively recruited back to the dereplicated MAG library using Bowtie2 (v2.4.2) (11) with the default minimum score threshold. This data set allows for comparisons of closely related microorganisms from different salinity zones and increases our understanding of estuarine pelagic microbial communities. These samples have corresponding 16S rRNA amplicon libraries (12).

Thanks go to Julian Damashek for sample collection and Linta Reji for advice on metagenomic workflows. Thanks go to Jim Cloern and the Water Quality of San Francisco Bay monitoring group at USGS and the R/V Polaris crew for facilitating our participation in numerous cruises. Some of the computing for this project was performed on the Stanford University Sherlock cluster. We would like to thank Stanford University and the Stanford Research Computing Center for providing computational resources and support that contributed to these research results.

This work was supported in part by NSF CAREER grant OCE-0847266 from the Biological Oceanography program (to C.A.F.) and in part by fellowship support from the NSF GRFP and Stanford Data Science Scholars program (to A.N.R.). The work conducted by the U.S. Department of Energy Joint Genome Institute, a DOE Office of Science User Facility, is supported by the Office of Science of the U.S. Department of Energy under Contract no. DE-AC02-05CH11231. Sequencing thanks to JGI CSP project 503022 to C.A.F.