Research map / Probabilistic and scientific modeling / Molecules and proteins
Doloris: Dual Conditional Diffusion Implicit Bridges with Sparsity Masking Strategy for Unpaired Single-Cell Perturbation Estimation
Changxi Chi, Jun Xia, Yufei Huang, Zhuoli Ouyang, Cheng Tan, Yunfan Liu, Jingbo Zhou, Chang Yu, Liangyu Yuan, Siyuan Li, Zelin Zang, Stan Z. Li
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 |
|---|---|---|---|
| use Doloris instead of the linear baseline | Doloris · Adamson unseen single-gene (all genes) | RMSE: 0.0473 -> 0.0336 | Improved |
| use Doloris instead of chemCPA | Doloris · sci-Plex3 unseen molecular perturbation (all genes) | RMSE: 0.0570 -> 0.0287 | Improved |
| use Doloris instead of scGPT | Doloris · Adamson unseen single-gene (all genes) | RMSE: 0.5372 -> 0.0336 | Improved |
| use Doloris instead of linear on double-gene Norman | Doloris · Norman double gene (all genes) | RMSE: 0.0405 -> 0.0385 | Improved |
| use Doloris instead of chemCPA on OOD drugs | Doloris · sci-Plex3 OOD drug (all genes) | RMSE: 0.0689 -> 0.0547 | Improved |
| add unperturbed-group mean and variance as model input (ablation) | Doloris · single-cell perturbation ablation | generation quality | Improved |
| initialize sampling from noised unperturbed-cell latent instead of random Gaussian noise (ablation) | Doloris · single-cell perturbation ablation | generation quality | Improved |
| add the mask model that predicts zero-valued genes (ablation) | Doloris · single-cell perturbation ablation | generation quality | Improved |
| increase diffusion sampling steps from 1 to 30 (ablation) | Doloris · NFE study | reconstruction metrics | Improved |
| increase diffusion sampling steps from 30 to 50 or 70 (ablation) | Doloris · NFE study | reconstruction metrics | No clear effect |
Closest papers by meaning
- PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis (NeurIPS 2025 · similarity 0.88)
- A benchmark for prediction of transcriptomic responses to chemical perturbations across cell types (NeurIPS 2024 · similarity 0.87)
- A Simple and Comprehensive Benchmark for Single-Cell Transcriptomics (AAAI 2025 · similarity 0.87)
- PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimation (NeurIPS 2025 · similarity 0.87)
- PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction (ICML 2025 · similarity 0.86)
- Cradle-VAE: Enhancing Single-Cell Gene Perturbation Modeling with Counterfactual Reasoning-based Artifact Disentanglement (AAAI 2025 · similarity 0.85)
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.
