Latitude- and depth-driven divergence in protist trophic strategies revealed by a machine learning model

Protists are ubiquitous across the ocean, holding different roles in the food web depending on their trophic capabilities. Many protists are mixotrophs, which are capable of both photosynthesizing and ingesting prey. However, there is limited knowledge of which protist species are mixotrophs in nature, as well as their activity and distribution throughout the ocean. Here, we present Marine PRotist <i>In Situ</i> trophic Mode predictor (MarPRISM), a refined XGBoost-based machine learning model for predicting the <i>in situ</i> trophic mode (phototrophy, mixotrophy, and heterotrophy) of marine protist species based on transcriptional profiles. We used MarPRISM to generate 1,462 trophic mode predictions for 28 environmental protist species based on 335 metatranscriptomes collected across the North Pacific Ocean, from the surface to 130 m depth, over the diel cycle, and within nutrient-amended incubations. Eight environmental species were identified as having mixotrophic capabilities, including six dinoflagellates, one bolidophyte, and one haptophyte. The species with mixotrophic capabilities varied in how they shifted their trophic mode across the surface ocean and in response to the experimental amendment of nitrate and iron. Limited light availability appeared to lead one species to shift from mixotrophy at the surface toward heterotrophy between 41 and 130 m depth. We used transcript abundance to evaluate the abundance of species with different trophic capabilities (species with mixotrophic capabilities, phototrophic specialists, and heterotrophic specialists). At the surface within the subtropical gyre, transcript abundance was similar among protist species with different trophic capabilities. In the gyre, the protist community was nitrate-limited, and experimental nitrate amendment favored phototrophic specialists. Increasing nitrate availability with latitude was correlated with phototrophic specialists being the dominant protist trophic group in the transition zone between the subtropical and subpolar gyres under high nitrogen availability. In contrast, under lower nitrogen conditions in the transition zone, protist species with different trophic capabilities comprised equal portions of the surface community. Light and nitrate availability influenced the transcript abundance of phototrophic specialists across depth; phototrophic specialists had high transcript abundance at 130 m in the subtropical gyre and at the surface in the transition zone, while species with mixotrophic capabilities and heterotrophic specialists showed less variation in transcript abundance with depth.

Authors

Institutions

Publication Details

Journal
Frontiers in Microbiology
Published
2025-09-01
DOI
https://doi.org/10.3389/fmicb.2025.1602162
Citations
2
Primary Topic
Microbial Community Ecology and Physiology
Type
article
Field-Weighted Citation Impact
3.43

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Latitude- and depth-driven divergence in protist trophic strategies revealed by a machine learning model

Elaina Thomas, Nicholas J. Hawco, Sacha Coesel, Randelle M. Bundy et al.
2 citations
Frontiers in Microbiology
Microbial Community Ecology and Physiology
3.43
article

Latitude- and depth-driven divergence in protist trophic strategies revealed by a machine learning model

Elaina Thomas, Nicholas J. Hawco, Sacha Coesel, Randelle M. Bundy, Ryan D. Groussman, E. Virginia Armbrust
article en
2 citations

Abstract

Protists are ubiquitous across the ocean, holding different roles in the food web depending on their trophic capabilities. Many protists are mixotrophs, which are capable of both photosynthesizing and ingesting prey. However, there is limited knowledge of which protist species are mixotrophs in nature, as well as their activity and distribution throughout the ocean. Here, we present Marine PRotist <i>In Situ</i> trophic Mode predictor (MarPRISM), a refined XGBoost-based machine learning model for predicting the <i>in situ</i> trophic mode (phototrophy, mixotrophy, and heterotrophy) of marine protist species based on transcriptional profiles. We used MarPRISM to generate 1,462 trophic mode predictions for 28 environmental protist species based on 335 metatranscriptomes collected across the North Pacific Ocean, from the surface to 130 m depth, over the diel cycle, and within nutrient-amended incubations. Eight environmental species were identified as having mixotrophic capabilities, including six dinoflagellates, one bolidophyte, and one haptophyte. The species with mixotrophic capabilities varied in how they shifted their trophic mode across the surface ocean and in response to the experimental amendment of nitrate and iron. Limited light availability appeared to lead one species to shift from mixotrophy at the surface toward heterotrophy between 41 and 130 m depth. We used transcript abundance to evaluate the abundance of species with different trophic capabilities (species with mixotrophic capabilities, phototrophic specialists, and heterotrophic specialists). At the surface within the subtropical gyre, transcript abundance was similar among protist species with different trophic capabilities. In the gyre, the protist community was nitrate-limited, and experimental nitrate amendment favored phototrophic specialists. Increasing nitrate availability with latitude was correlated with phototrophic specialists being the dominant protist trophic group in the transition zone between the subtropical and subpolar gyres under high nitrogen availability. In contrast, under lower nitrogen conditions in the transition zone, protist species with different trophic capabilities comprised equal portions of the surface community. Light and nitrate availability influenced the transcript abundance of phototrophic specialists across depth; phototrophic specialists had high transcript abundance at 130 m in the subtropical gyre and at the surface in the transition zone, while species with mixotrophic capabilities and heterotrophic specialists showed less variation in transcript abundance with depth.

Frontiers in MicrobiologyVol. 16
University of Hawaiʻi at Mānoa (US), University of Washington (US), University of Hawaiʻi Sea Grant (US)
National Science Foundation
Openalex Percentile: Top 8%
Microbial Community Ecology and Physiology
3.43
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.