ML (VAE, custom CLADES) delimits second Micrognathozoa species

AI system

System: Machine-learning species delimitation on single-specimen transcriptomes: unsupervised clustering (PCA, random forest, t-SNE, variational autoencoder) and supervised classifiers (delimitR, CLADES)
Model: Not assessed
Release: Not assessed
Provider: Not assessed
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Field collection of all three known populations. Single-specimen transcriptomics for microscopic animals. Species-delimitation models run alongside conventional phylogenetics.
Human oversight: Not assessed
Relation to AI: The paper's title credits machine learning with the identification. Each machine-learning method run (unsupervised PCA, random forest, t-SNE and VAE; supervised delimitR and CLADES) separated the Crozet population. The adopted two-species result rests on the authors' most conservative estimate across PTP, VAE and custom CLADES. It does not rest on delimitR's three-species model, which would also have split Greenland from the Pyrenees. The conclusion also rests on deep phylogenetic divergence, extreme geographical separation and differences in COI and 18S rRNA sequences. So the machine-learning output was one of several lines of evidence, not the sole basis.
AI contribution: The paper's title credits machine learning with the identification. Each machine-learning method run (unsupervised PCA, random forest, t-SNE and VAE; supervised delimitR and CLADES) separated the Crozet population. The adopted two-species result rests on the authors' most conservative estimate across PTP, VAE and custom CLADES. It does not rest on delimitR's three-species model, which would also have split Greenland from the Pyrenees. The conclusion also rests on deep phylogenetic divergence, extreme geographical separation and differences in COI and 18S rRNA sequences. So the machine-learning output was one of several lines of evidence, not the sole basis.

Benefits

Outcome identifiers: Not assessed
Benefit: Transcriptomes plus machine-learning delimitation split the Crozet population as Limnognathia desmeti sp. nov. (Proc R Soc B, 19 February 2025).
Category: Improved Knowledge Discovery
Status: Not assessed
Beneficiary groups: People. Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Measurement

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: Not assessed
Comparison: Not assessed
Results: Not assessed
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Money

Kind of value: Not assessed
Amount: Not assessed
Currency and prices: Not assessed
Period and unit: Not assessed
Realization: Not assessed
Recipients and payers: Not assessed
Gross or net: Not assessed
Included costs: Not assessed
Comparison: Not assessed
Method: Not assessed
Return and payback: Not assessed

People and places

Participants: Operator: University of Copenhagen Marine Biological Section and co-author institutions.
Jurisdictions: Benefit: Pointe du Bougainville, Île de la Possession, Crozet Archipelago (Subantarctic). Operator: Denmark.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Follow-up

Conditions for replication: Sufficient genomic signal between populations. Agreement between machine learning (ML) and standard delimitation. Morphological comparison to document cryptic status.
Replication attempts: 1 follow-up entries. Benefit did not recur: 1.
Later findings: The Crozet population was formally named Limnognathia desmeti sp. nov. in the same February 2025 paper, and no later development was found.
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Evidence

Source numbers: AIBD-0106-source-0001–0002
Citations: Numbered references below.
Language: Not assessed
Source relationships: Operator
Original evidence: Not assessed
Access and versions: Not assessed
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Grades

Assessed benefits: People. Organizations. Society.
Depth: People: 1. Organizations: 1. Society: 1.
Reasons: People: A verified discrete result for the research team: formal description of a new species in a phylum with one previously known species. Organizations: Adds a named taxon and a Subantarctic record to museum and university collections; no operational change documented. Society: Knowledge gain about a rare phylum's diversity and dispersal; no conservation action or beneficiary population is documented.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: The deployer's account

Entry details

Kind: Not assessed
Related entries: Not assessed
Field: Conservation and biodiversity. Taxonomy and biodiversity discovery.
Purpose: Test species hypotheses within Micrognathozoa using transcriptomic data and machine-learning species delimitation.
Dates: Date basis unassessed: 2025 (precision unassessed). Updated: 21 October 2026.
Stage: Not assessed
Version: SHA-256: ae943d69d3df…

References

Other identifiers: Not assessed
Category mappings: Not assessed
Brief: Not assessed
Revisions: 4 revisions. Latest: 21 October 2026.

Sources

Peer-reviewed study Deployer

Sato, S., Appeldorff, C., Wangensteen, O. S., Garcés-Pastor, S., Laumer, C. E., Herranz, M., Giribet, G., Renault, D., Møller, P. R., & Worsaae, K. (2025). Phylogenomics of the rarest animals: a second species of Micrognathozoa identified by machine learning. Proceedings of the Royal Society B 292(2041). doi:10.1098/rspb.2024.2867
Peer-reviewed study Deployer

Sato, S., et al. (2025). Phylogenomics of the rarest animals: a second species of Micrognathozoa identified by machine learning. Proc R Soc B (full text, PMC11836703).