analiza aiqbench
This paper applies representation learning to microbial genome classification, aiming to identify species absent from training data by aligning genome encodings with a taxonomy-derived embedding space. The reported gains over pseudo-image and transformer-graph baselines are concentrated at higher taxonomic ranks, and the results come from benchmark and disjoint-species evaluations that the authors present as a complementary candidate-generation tool rather than a replacement for reference-based methods. For ML readers, the useful takeaway is the zero-shot and open-set design, though the claims need independent validation on real-world sequencing data.
HyperTaxa: Multiscale Genome Representations for Generalized Microbial Taxonomic Inference
arXiv:2610.25673v1 Announce Type: new Abstract: 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.
Abstract: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.
Artificial Intelligence (cs.AI); Genomics (q-bio.GN)
arXiv:2610.25673 [cs.AI]
(or arXiv:2610.25673v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.25673 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 271 cuvinte
Subiecte conexe
The Role of Fine-grained Harm Signals in LLM Safety
arXiv · 18 sept. 2026
The Generalized BEADS Representation: An Informationally Complete Visualization of Mixed Quantum States
arXiv · 15 oct. 2026
Generalized PINN-Based Inverse Estimation of Spatially Varying Heat Transfer Coefficients with Transfer Learning
arXiv · 14 oct. 2026
Differentiated Aggregation to Improve Generalization in Federated Learning
arXiv · 27 aug. 2026