Researchers have developed SCALE, a novel conditional transport model designed to predict cellular responses to perturbations. Unlike previous methods that required paired control and treated cell populations, SCALE can learn perturbation-specific effects from unpaired data. The model utilizes a shared set-aware encoder and a conditional Diffusion Transformer backbone to predict treated cell populations, demonstrating effectiveness across various genetic, chemical, developmental, and immune perturbations. In experiments with CRISPR data, SCALE outperformed existing methods in predicting gene-expression changes and response directions, and it successfully prioritized cytokines likely to induce distinct immune responses. AI
IMPACT Enables more efficient and accurate prediction of cellular responses to various treatments, potentially accelerating biological research and drug discovery.
RANK_REASON The cluster contains a research paper detailing a new AI model for biological predictions. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- CRISPR
- DagsHub
- Diffusion Transformer
- Hugging Face
- peripheral blood mononuclear cell
- SCALE
- Shuizhou Chen
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
