Abstract
Climate change is increasingly affecting Pacific salmon to combinations of thermal, hypoxic, and osmotic stressors that can impair survival. Although genomic tools have enabled detailed insight into stressor responses under controlled conditions, field applications remain limited by the lack of species-specific classifiers capable of resolving physiological condition across multiple stressor contexts. In this study, we developed and experimentally validated transcriptional classifiers for juvenile sockeye salmon (Oncorhynchus nerka) using controlled multi-stressor challenges. Fish exposed to a full factorial design of three temperatures (10 °C, 14 °C, 18 °C), three salinities (0, 20, 28 ppt), and two dissolved oxygen levels (> 8 mg L-1, 3–3.5 mg L-1) over six days, across smolt and de-smolt stages; pre-smolts were additionally challenged with seawater (28 ppt). Gill tissue, a primary interface between fish and their environment, was analyzed using the “Salmon Fit-Chips” microfluidic qPCR tool to quantify gene expression responses. Random Forest models were then used to develop classifiers detecting transcriptional signatures of thermal stress, hypoxic stress, salinity acclimation, and smolt stage. Classifiers achieved high predictive accuracy (85.5%–100%) across all stressors and life stages. Recovery analysis showed that transcriptional stress signals diminished within three days for hypoxia, but lingered longer for thermal stress. By developing sockeye salmon–specific molecular classifiers within a multi-stressor experimental framework, this study advances the Fit-Chip approach and provides a foundation for future field applications, including potential non-lethal monitoring using gill biopsies.
Similar content being viewed by others
Molecular signatures and machine learning driven stress biomarkers for rainbow trout aquaculture and climate adaptation
Linking aerobic scope to fitness in the wild reveals potential opportunities to help recover imperiled salmon populations
Comparative study of fish adaptation at cellular and genetic levels across eco-sensitive zones with contrasting oxygen and temperature regimes
Acknowledgements
We thank DFO Aquarium services (Christine MacWilliams, Holly Hicklin, Ted Sweeten, Allison Latell, and Carson Jewitt) for their help in the design and assembly of the experiment. We also thank Dr. Christoph Deeg and coop students, Amanda Silveri, Rachel Witt, Jennifer Liu, Loclan, and the biologists Jenna Fleet and BryanQiuwen Ding for their help with sample collection.
Funding
This research was funded by the Fisheries and Oceans Canada (DFO) under the CSRF fund (BG-01–01) and also supported by the Pacific Salmon Foundation (PSF).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
41598_2026_49226_MOESM1_ESM.pdf (download PDF )
Supplementary Information 1: Canonical plots of the first two principal components of the 8 optimal biomarkers for hypoxic stress. Principal component analysis was performed on the 2/3 training set (A) and then tested to the 1/3 testing set (B). PCA was also applied on the 3 days recovery from hypoxia to normoxia set (C). Ellipses represent 95% confidence areas for the groups within treatments using the training set; centroids are represented by the largest point of the same colour. Arrows (left panel) represent loading vectors of the biomarkers using the training set.
41598_2026_49226_MOESM2_ESM.pdf (download PDF )
Supplementary Information 2: Canonical plots of the first two principal components of the five optimal biomarkers for salinity acclimation. Principal component analysis was performed on the 2/3 training set (A) and then tested to the 1/3 testing set (B) using the freshwater, brackish water and seawater data. Ellipses represent 95% confidence areas for the groups within treatments using the training set; centroids are represented by the largest point of the same colour. Arrows (left panel) represent loading vectors of the biomarkers using the training set.
41598_2026_49226_MOESM3_ESM.pdf (download PDF )
Supplementary Information 3: Canonical plots of the first two principal components of the 11 optimal biomarkers for smolt stage detection. Principal component analysis was performed on the 2/3 training set (A) and then tested to the 1/3 testing set (B). Ellipses represent 95% confidence areas for the groups within treatments using the training set; centroids are represented by the largest point of the same colour. Arrows (left panel) represent loading vectors of the biomarkers using the training set.
Supplementary Information 4. (download DOCX )
Supplementary Information 5. (download DOCX )
Supplementary Information 6. (download DOCX )
Supplementary Information 7. (download DOCX )
Supplementary Information 8. (download DOCX )
Supplementary Information 9. (download DOCX )
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
About this article
Cite this article
Akbarzadeh, A., Ming, T., Li, S. et al. Developing molecular classifiers to detect environmental stressors and smolt stages in sockeye salmon, Oncorhynchus nerka.
Sci Rep (2026). https://doi.org/10.1038/s41598-026-49226-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-49226-3
Keywords
- Salmon
- Stressors
- Biomarker
- Random forest
- Fit-chips
- Smolt
- De-smolt
- Climate change
Source: Ecology - nature.com
