Table of Contents

• Coverage
• Dataset Description
  • Methods & Sampling
  • Data Processing Description
  • BCO-DMO Processing Description
• Data Files
• Related Publications
• Related Datasets
• Parameters
• Instruments
• Deployments
• Project Information
• Funding

Related data table and dataset descriptions:

Total spectral counts refer to the total number of spectra with peptide to spectrum matches (PSMs) that matches to each entry within the FASTA sequence database. This approach allows each peptide to map to multiple closely related sequences. In contrast, with exclusive spectral counts each peptide is only allowed to map to one sequence within the FASTA database, and when a peptide is found in multiple database sequences the one with the most peptides mapping (parsimony) to it is selected. There are pros and cons to each approach, where total spectral counts will double count peptides when two similar proteins are compared, and exclusive spectral counts will underrepresent less abundant proteins with shared peptides, favoring the most homolog with the most shared peptides. Considering protein groups with shared peptides or focusing on peptide-level analyses are alternative approaches that could be constructed from these results.  

See "Related Datasets" section for:
*  "AE1913 Protein Spectral Counts" which includes the exclusive and total spectral counts associated with proteins.
*  "AE1913 Peptide Spectral Counts" which includes the individual peptides associated with these proteins (includes total spectral counts for each peptide).

CTD and other data from the same cruise are listed on deployment page AE1913: https://www.bco-dmo.org/deployment/916412

These data will become part of the Ocean Protein Portal (https://proteinportal.whoi.edu/; Saito et al., 2020).

The assembly, annotations, metatranscriptomic assembly products, the same exclusive protein spectral counts, and other useful information associated with this multi-omic analysis was published as a package at Zenodo (doi: 10.5281/zenodo.8287779). 

Methods are reported in Cohen et al. 2023 (biorxiv preprint doi: 10.1101/2023.11.20.567900) and are summarized below.
* This section describes how this and related datasets were generated (see "Related Datasets" section).

One half of the 142 mm filters (0.2-51 μm) collected by Clio were processed for metaproteomics. Proteins were extracted in an 1% SDS-based detergent in 50 mM HEPES at pH 8.5, reduced with dithiothreitol, alkylated with iodoacetamide, and purified using a polyacrylamide electrophoresis tube gel method. Protein quantification was performed using a BSA assay. Trypsin was added to the protein-bead mixture in a 1:20 trypsin:protein ratio. Peptides were purified using C18 tips and diluted to a concentration of 0.1 μg μL−1. 

Approximately 2-5 µg of purified peptides were injected onto a Dionex UltiMate 3000 RSLCnano LC system with an additional RSLCnano pump, run in online 2D active modulation mode interfaced with a Thermo Fusion mass spectrometer. The mass spectrometer acquired MS1 scans from 380 to 1,580 m/z at 240K resolution in the Orbitrap. MS2 were collected in data dependent mode in the ion trap with a cycle time of 2 seconds between scans and acquisition of charge states 2 to 10. MS2 scans had 1.6 m/z isolation window, 50 ms maximum injection time and 5 s dynamic exclusion time. 

The metatranscriptomic ORFs were used as the protein database, and peptide-spectrum matches were performed using Sequest algorithm within IseNode Proteome Discoverer 2.2.0.388 with a parent ion tolerance of 10 ppm and fragment tolerance of 0.6 Da, and 0 max missed cleavage. Identification criteria consisted of a peptide threshold of 98% (false discovery rate [FDR] = 0.1%) and protein threshold of 99% (1 peptide minimum, FDR = 1.5%) in Scaffold 5.1.2 (Proteome Software) resulting in 77,438 proteins and 3,155,061 exclusive spectral counts. 

No additional processing besides extracting from a package .zip and renaming filename .fasta for clarity. Confirmed with the author it is in FASTA format.
* Downloaded from part of Zenodo v2 package (doi: https://zenodo.org/records/8287779) file
merged.fasta.transdecoder.pep.zip and the filename was "merged.fasta.transdecoder.pep" renamed "merged_fasta_transdecoder_pep.fasta"

NSF Award Abstract:
Microscopic communities in the ocean can be surprisingly diverse. This diversity makes it difficult to study the individual organisms and reactions that control specific reactions controlling nutrient cycles. Past studies confirm that iron and nitrogen are vital elements for biological growth. There is increasing evidence, however, that other chemicals such as silica, zinc, cobalt, and vitamin B12 may be just as important. This project will provide an unprecedented view of community distributions using new molecular methods to isolate and link active proteins to specific chemical cycles during the very first research deployment of a brand-new autonomous underwater vehicle (AUV). The AUV will collect samples in programed patterns by pumping water directly into its filtering mechanism and then return the samples to the ship for analysis. The Bermuda Atlantic Time-series Study (BATS) station, which provides abundant supporting data, is the site for this innovative investigation into the microbial ecology and chemistry of the open oceans. Additionally, data will be widely distributed to other scientists through the Ocean Protein Portal website being developed by the Woods Hole Oceanographic Institute (WHOI) and the Biological and Chemical Oceanography Data Management Office. Data will also contribute a new teaching module in the Marine Bioinorganic Chemistry course at WHOI.

This first scientific deployment of the newly engineered and constructed biogeochemical AUV, Clio, will generate a novel dataset to examine marine microbial biogeochemical cycles in the Northwestern Atlantic oligotrophic ocean in unprecedented detail and at high vertical resolution. First the project proposes to understand if the microbial community reflects the varying chemical composition and cyanobacterial species through nutrient response adaptations. Additionally, the research will determine if iron stress in the low light Prochlorococcus ecotyope found in the deep chlorophyll maximum is a persistent feature influenced by seasonal dust fluxes. The highly resolved vertical data from the in situ pumping capabilities of Clio are fundamental to a rigorous examination of these biogeochemical questions. This highly transformative dataset will greatly advance understanding of the nutrient and trace element cycling of this region and will be the first field validation of the potentially revolutionary capability these new approaches represent for the study of marine microbial biogeochemistry.