Deep-sea cold seep stage assessment has traditionally relied on costly, high-risk manned submersible operations and visual surveys of macrofauna. Although microbial communities provide a promising and more cost-effective alternative, reliable inference remains challenging because the available deep-sea dataset is extremely small (n=13) relative to the microbial feature dimension (p=26), making purely data-driven models highly prone to overfitting. To address this, we propose a knowledge-enhanced classification framework that incorporates an ecological knowledge graph as a structural prior. By fusing macro-microbe coupling and microbial co-occurrence patterns, the framework internalizes established ecological logic into a Graph-Regularized Multinomial Logistic Regression (GRMLR) model, effectively constraining the feature space through a manifold penalty to ensure biologically consistent classification. Importantly, the framework removes the need for macrofauna observations at inference time: macro-microbe associations are used only to guide training, whereas prediction relies solely on microbial abundance profiles. Experimental results demonstrate that our approach significantly outperforms standard baselines, highlighting its potential as a robust and scalable framework for deep-sea ecological assessment.
Figures & tables
Figure 1
Figure 3: Motivation: Cold seep development stage recognition requires robust learning from complex microbial data, since macrofaunal observations are expensive and challenging to acquire.
Figure 4: Framework of GRMLR and data flow. Macrofauna counts and CLR-transformed microbial features are integrated through an ecological knowledge graph during training, while inference uses only microbial data for stage classification.
Figure 4
Statistic
Value
No. of sampling sites n
13
No. of microbial taxa p
26
No. of developmental stages K
3
Stage distribution
Juvenile nJ=3 / Adult nA=7 / Dead nD=3
No. of macrofauna categories
4 (dead, adult, juvenile, Calyptogena )
Microbial taxa ( p=26 , class-level):
Table 1: Dataset statistics of the cold seep sampling sites.
Method
Accuracy
Macro-F1
Stage-wise correct
Juvenile
Adult
Dead
LR (CLR + L2)
0.6154
0.6000
2
5
1
SVM-RBF (CLR)
0.6154
0.5794
1
5
2
Random Forest (CLR)
0.4615
0.3072
0
5
1
KNN k=3 (CLR)
0.3846
0.1852
0
5
0
LR (Raw)
0.4615
0.4222
1
4
1
Table 2: LOOCV classification results on the cold seep dataset. The last three columns report the number of correctly predicted samples for each developmental stage (Juvenile: nJ=3 ; Adult: nA=7 ; Dead: nD=3 ). The best results are shown in Bold , while the second-best results are Underlined .
Figure 7: (a) Permutation test null distribution for GRMLR ( p=0.02 ); the dashed line marks the observed accuracy of 84.62%. (b) Ablation study results; each point shows accuracy and Macro-F1 when the corresponding component is removed.
Figure 8: (a) Sensitivity to the graph mixing parameter α . The shaded region marks theoptimal range α∈[0.1,0.9] where both graph components actively contribute. (b) Top-10 microbial taxa by GRMLR coefficient magnitude.
Microbiome functions are encoded within the genes of the community-wide metagenome. A natural question is whether properties of a microbial community can be predicted just from knowing the raw DNA sequences of its members. In this work, we employ set-aggregated genome embeddings (SAGE) to predict community-level abundance profiles, exploiting the few-shot learning capabilities of genomic language models (GLMs). We benchmark this approach to show improved generalization on novel genomes compared to classical bioinformatics approaches. Model ablation shows that community-level latent representations directly result in improved performance. Lastly, we demonstrate the benefits of intermediate transformations between latent representations and demonstrate the differences between GLM embedding choices.
Younhun Kim, Georg K. Gerber, Travis E. Gibson
1Brigham and Women’s Hospital, Boston, MA, USA · 2Harvard Medical School, Boston, MA, USA
Automated classification of marine species from underwater imagery is essential for scalable ocean biodiversity monitoring and conservation policy. Existing approaches struggle with severe domain shift across collection platforms, fine-grained visual similarity between closely related species, and uneven annotation granularity, where many specimens can only be identified to genus or a coarser taxonomic rank. We present a taxonomy-aware deep learning framework that aligns both the training loss and the inference rule with the hierarchical structure of biological classification, combining a taxonomy-weighted loss, minimum-risk Bayesian inference, multi-scale feature encoding, and independent per-rank classification heads. Evaluated on the FathomNet 2025 dataset1 (79 marine classes across seven taxonomic ranks), the system achieves a mean taxonomic distance of 1.581, within 3% of the 1st-place solution (1.535), with the largest gains from metric-aligned inference and simple, decoupled components that generalize better than learned dependencies under distribution shift.
Dan Zimmerman, Dimitris A. Pados, George Sklivanitis
Florida Atlantic University, Boca Raton, FL 33431, USA
Reliable microbial taxonomic inference is challenged by the rapid growth of newly sequenced organisms and incomplete reference coverage. While alignment- and database-driven tools remain essential for reference-backed identification, they are not designed for learned inference over species absent from genome-level training. We introduce HyperTaxa, a taxonomy-aligned representation learning framework for open-set microbial taxonomic inference. HyperTaxa encodes whole genomes as multiscale hyperspectral k-mer transition fields and aligns these representations to a taxonomy-derived semantic space using a two-stage Node2Vec-to-MDS procedure that combines local graph structure with global hierarchical distance preservation. The resulting space supports zero-shot inference over species excluded from genome-level training but represented in the taxonomy, generalized zero-shot inference over seen and unseen species, and cascaded genus-to-species prediction. Across benchmark and large-scale disjoint-species evaluations, HyperTaxa improves taxonomic prediction over pseudo-image and transformer-graph baselines, with the strongest gains at higher taxonomic ranks. Ablations further show that multiscale transition encoding and taxonomy-aware alignment are key contributors to generalization. These results position HyperTaxa as a complementary taxonomy-aware representation and candidate-generation framework for microbial identification under incomplete reference coverage and open-set uncertainty.
Akash Rao, Sushim Gupta, Vishalini Ramnath +2
CSSE Department, Auburn University, Auburn, AL, USA · Oklahoma Animal Disease Diagnostic Laboratory, Oklahoma State University, Stillwater, OK, USA