arXiv:2605.24370v1 Announce Type: new Abstract: Behavioral phenotyping of genetic animal models currently requires labor-intensive manual feature engineering that limits reproducibility and scalability. We present GEESE, an end-to-end deep learning framework that learns behavioral representations directly from 3D pose dynamics without hand-crafted features. Using a pretrained time series foundation model, we encode movement sequences into a behavioral manifold that supports both behavior classification and genotype prediction. Evaluated across three autism-associated genetic models (CNTNAP2, C
Advances in deep learning and 3D pose estimation are converging, enabling more sophisticated and automated analysis of complex biological data.
This development significantly enhances the precision and scalability of behavioral phenotyping in genetic models, crucial for understanding complex neurological disorders and developing targeted interventions.
The ability to automatically learn behavioral representations from raw data removes a major bottleneck in genetic research by eliminating labor-intensive manual feature engineering.
- · Neuroscience researchers
- · Pharmaceutical companies
- · AI in biology startups
- · Traditional manual phenotyping labs
Accelerated discovery of genetic links to behavioral phenotypes in animal models.
Improved understanding and more precise diagnostics for human neurological disorders like autism.
Potential for new therapeutic targets and personalized medicine approaches based on deep behavioral insights.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG