TL;DR
This paper establishes a standardized, consensus-based vocabulary for the 13 most common marine microbial groups identified via flow cytometry (FCM). By providing a logic-based decision tree and machine-readable identifiers, the authors aim to transform fragmented oceanographic data into a "FAIR" (Findable, Accessible, Interoperable, and Reusable) global resource for climate and ecosystem modeling.
What Problem Does This Paper Address?
Marine microorganisms—from tiny Prochlorococcus to larger nanophytoplankton—are the engines of the ocean’s biogeochemical cycles, producing half the Earth's oxygen and forming the base of the food web. Since the 1980s, Flow Cytometry (FCM) has been the gold standard for studying these cells at a single-cell level, recording their light-scattering and fluorescence properties in real-time.
However, the field has long suffered from a "Tower of Babel" problem. Because different researchers use different instrument settings and naming conventions, one lab’s "picoeukaryote" might be another’s "RedPico." This lack of standardization makes it nearly impossible to combine datasets from different cruises or decades to perform large-scale global analyses. As autonomous sensors begin to generate massive amounts of data, the need for a machine-readable, interoperable language has become critical.
Methods
The authors developed a consensus framework through extensive consultation with FCM experts and a literature review spanning 1983–2020. Key components include:
- Consensus Vocabulary: A list of 13 microbial groups defined by their optical signatures (e.g., chlorophyll-a red fluorescence, phycoerythrin orange fluorescence, and light scatter).
- Standardized Cytograms: A common layout for 2D log-transformed plots to visualize cell populations consistently.
- Classification Decision Tree: A logic-based flow (Figure 2 in the paper) that guides researchers through the identification process based on specific fluorescence and scattering thresholds.
- Technical Governance: The vocabulary is hosted on the NERC Vocabulary Server (NVS) using the Simple Knowledge Organization System (SKOS). Each term is assigned a unique, permanent URI (Uniform Resource Identifier), making the definitions machine-readable and "future-proof."
Datasets
While the paper does not introduce a new biological dataset, it is built upon:
- Annex 1 (Supplementary Material): A comprehensive extraction of terms and group definitions from the literature between 1983 and 2020.
- Reference Standards: The use of 3 µm silica beads as a standardized internal proxy to separate "pico" (< 2–3 µm) from "nano" (3–20 µm) size classes.
Key Insights and Ideas
- Moving Beyond Taxonomy: The paper acknowledges that FCM rarely provides true species-level taxonomy. Instead, it focuses on Functional Groups, which are often more relevant for ecological modeling than precise species names.
- The 3µm Pivot: Traditionally, the "pico-nano" cutoff was defined by 2 µm filters. The authors advocate for using 3 µm silica beads as a more robust optical standard, as cell size can fluctuate significantly during the cell cycle.
- FAIRification is a Social Process: The authors emphasize that technical standards are useless without community buy-in. By hosting the vocabulary on GitHub and the NVS, they allow for transparent, community-driven updates.
- Relative vs. Absolute: Because FCM light units are arbitrary and vary by instrument, the vocabulary relies on the relative positions of groups on a cytogram rather than absolute numerical values.
Key Results
- The "Look-up Table" (Table 1): Provides a definitive map between common "jargon" terms (like Syn or HNA) and standardized labels (like OraPicoProk or HetHNA).
- Machine-Readability: Every group is now linked to a URI (e.g.,
SDN:F02::F0200002for RedPicoProk). This allows data repositories to automatically cross-reference datasets. - The Decision Tree: Provides a standardized "if-then" logic for classifying particles, which is essential for training future machine learning and convolutional neural network (CNN) models for automated ocean monitoring.
Limitations and Future Work
- Instrument Variability: The authors admit that while the relative positions of groups are consistent, the absolute "brightness" of a cell depends heavily on instrument settings and laser power.
- Excluded Groups: This initial vocabulary excludes viruses and microzooplankton, as they are not yet routinely analyzed using the specific automated, unstained methods discussed here.
- The "Expert" Gap: While the decision tree helps, some groups (like those with overlapping fluorescence) still require expert manual "gating."
- Future Directions: The authors suggest expanding the vocabulary to include imaging-in-flow data (which adds cell shape/morphology) and more complex staining protocols for heterotrophic organisms.
Summary updated