Research map / Probabilistic and scientific modeling / Molecules and proteins
Integrating Biological Knowledge for Robust Microscopy Image Profiling on De Novo Cell Lines
Jiayuan Chen, Thai-Hoang Pham, Yuanlong Wang, Ping Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add biological knowledge (graph regularization + scFM cell line tokens) | WSL · U2OS | top-1 accuracy: 4.12 -> 4.79 | Improved |
| add biological knowledge | SimCLR · VERO | top-1 accuracy: 32.82 -> 38.71 | Improved |
| add biological knowledge | MoCo v3 · VERO | top-1 accuracy: 20.73 -> 25.56 | Improved |
| add biological knowledge | MAE · HRCE | top-1 accuracy: 1.24 -> 1.79 | Improved |
| add biological knowledge | BYOL · U2OS | top-5 accuracy: 8.02 -> 8.05 | No clear effect |
| add cell line-specific representation (CS) alone (ablation) | WSL · U2OS | top-1 accuracy: 4.12 -> 4.64 | Improved |
| add perturbation-specific graph regularization (PS) alone (ablation) | WSL · U2OS | top-1 accuracy: 4.12 -> 4.25 | Improved |
| add both CS and PS (ablation) | WSL · U2OS | top-1 accuracy: 4.12 -> 4.79 | Improved |
| add cell line-specific representation (CS) alone (ablation) | BYOL · U2OS | top-5 accuracy: 8.02 -> 7.95 | Hurt |
| add perturbation relational graph (PS) alone (ablation) | BYOL · U2OS | top-1 accuracy: 3.61 -> 3.97 | Improved |
| increase scFM output dimensionality to 512 (vs 128 or 256) (ablation) | WSL · U2OS | top-1 accuracy | Hurt |
| use scGPT instead of scVI as scFM (ablation) | WSL · U2OS | top-1 accuracy | No clear effect |
Closest papers by meaning
- ViTally Consistent: Scaling Biological Representation Learning for Cell Microscopy (ICML 2025 · similarity 0.87)
- CellCLIP - Learning Perturbation Effects in Cell Painting via Text-Guided Contrastive Learning (NeurIPS 2025 · similarity 0.86)
- MorphoHELM: A Comprehensive Benchmark for Evaluating Representations for Microscopy-Based Morphology Assays (NeurIPS 2026 · similarity 0.85)
- A Cross Modal Knowledge Distillation & Data Augmentation Recipe for Improving Transcriptomics Representations through Morphological Features (ICML 2025 · similarity 0.85)
- Learning Identifiable Factorized Causal Representations of Cellular Responses (NeurIPS 2024 · similarity 0.85)
- PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
