Sky islands of Southwest China. II: Unraveling hidden species diversity of talpid moles using phylogenomics and skull-based deep learning
Bibliographic record
Abstract
Abstract The sky islands of Southwest China, characterized by dramatic topographical and climatic variations, are prominent hotspots of biodiversity and endemism. Organisms inhabiting middle-to-high elevation habitats in this region are geographically isolated within distinct mountain chains, which over geological time have been subjected to isolation-by-distance and isolation-by-environment. These processes have led to profound allopatric diversification and strong phylogeographic structuring, resulting in a plethora of genetically distinct cryptic species, as is becoming increasingly evident for many small mammal families. However, morphological conservatism can pose significant challenges in delineating these clades within species complexes. In this study, we leverage artificial intelligence technologies to unravel the hidden species diversity of moles (family Talpidae) in Southwest China’s sky islands. We first employed ultraconserved elements (UCEs) to investigate the evolutionary history of talpid moles, conducted molecular species delimitation using mitochondrial and multi-locus genes, and utilized both traditional and geometric morphometrics to examine their morphological disparity. To address the challenges of morphology based cryptic species identification, we developed a deep learning Hierarchical Identification of Species NETwork (HIS-NET) to create an image-based model that analyzes four different views of the skull/mandible to distinguish genera and species hierarchically. HIS-NET not only achieved expert-level accuracy in species identification but also effectively distinguished between cryptic and known species, aiding in the identification of key morphological variation intervals. Our results support the recognition of allopatrically distributed taxa in Euroscaptor and Parascaptor as full species, thereby confirming that species diversity in this region remains underestimated. Beyond advancing our understanding of speciation in this unique and fragile ecosystem, our study serves as a proof-of-concept, demonstrating the power of deep learning in unraveling hidden biodiversity within this and other species complexes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".