HarmonyCell: Automating Single-Cell Perturbation Modeling under Semantic and Distribution Shifts
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911495079591936 |
|---|---|
| author | Huang, Wenxuan Tsoi, Mingyu Huang, Yanhao Mao, Xinjie Xia, Xue Wu, Hao Wei, Jiaqi Yang, Yuejin Yu, Lang Tan, Cheng Zhang, Xiang Gao, Zhangyang Sun, Siqi |
| author_facet | Huang, Wenxuan Tsoi, Mingyu Huang, Yanhao Mao, Xinjie Xia, Xue Wu, Hao Wei, Jiaqi Yang, Yuejin Yu, Lang Tan, Cheng Zhang, Xiang Gao, Zhangyang Sun, Siqi |
| contents | Single-cell perturbation studies face dual heterogeneity bottlenecks: (i) semantic heterogeneity--identical biological concepts encoded under incompatible metadata schemas across datasets; and (ii) statistical heterogeneity--distribution shifts from biological variation demanding dataset-specific inductive biases. We propose HarmonyCell, an end-to-end agent framework resolving each challenge through a dedicated mechanism: an LLM-driven Semantic Unifier autonomously maps disparate metadata into a canonical interface without manual intervention; and an adaptive Monte Carlo Tree Search engine operates over a hierarchical action space to synthesize architectures with optimal statistical inductive biases for distribution shifts. Evaluated across diverse perturbation tasks under both semantic and distribution shifts, HarmonyCell achieves a 95% valid execution rate on heterogeneous input datasets (versus 0% for general agents) while matching or even exceeding expert-designed baselines in rigorous out-of-distribution evaluations. This dual-track orchestration enables scalable automatic virtual cell modeling without dataset-specific engineering. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01396 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HarmonyCell: Automating Single-Cell Perturbation Modeling under Semantic and Distribution Shifts Huang, Wenxuan Tsoi, Mingyu Huang, Yanhao Mao, Xinjie Xia, Xue Wu, Hao Wei, Jiaqi Yang, Yuejin Yu, Lang Tan, Cheng Zhang, Xiang Gao, Zhangyang Sun, Siqi Artificial Intelligence Computational Engineering, Finance, and Science Quantitative Methods Single-cell perturbation studies face dual heterogeneity bottlenecks: (i) semantic heterogeneity--identical biological concepts encoded under incompatible metadata schemas across datasets; and (ii) statistical heterogeneity--distribution shifts from biological variation demanding dataset-specific inductive biases. We propose HarmonyCell, an end-to-end agent framework resolving each challenge through a dedicated mechanism: an LLM-driven Semantic Unifier autonomously maps disparate metadata into a canonical interface without manual intervention; and an adaptive Monte Carlo Tree Search engine operates over a hierarchical action space to synthesize architectures with optimal statistical inductive biases for distribution shifts. Evaluated across diverse perturbation tasks under both semantic and distribution shifts, HarmonyCell achieves a 95% valid execution rate on heterogeneous input datasets (versus 0% for general agents) while matching or even exceeding expert-designed baselines in rigorous out-of-distribution evaluations. This dual-track orchestration enables scalable automatic virtual cell modeling without dataset-specific engineering. |
| title | HarmonyCell: Automating Single-Cell Perturbation Modeling under Semantic and Distribution Shifts |
| topic | Artificial Intelligence Computational Engineering, Finance, and Science Quantitative Methods |
| url | https://arxiv.org/abs/2603.01396 |