Research map / Graph learning / Knowledge and temporal graphs
Adaptive Data-Knowledge Alignment in Genetic Perturbation Prediction
Yuanfang Xiang, Lun Ai
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 neuro-symbolic alignment and refinement stages (A-R-A-R) (ablation) | GNN · E. coli unseen perturbations | balanced consistency: 0.3689 -> 0.4288 | Improved |
| add knowledge refinement after alignment (ablation) | ALIGNED (GNN) · E. coli | knowledge consistency: knowledge cons. 0.3714 -> 0.4520 | Improved |
| add alignment on top of MLP only (ablation) | MLP · E. coli | data consistency: data cons. 0.3312 -> 0.3872 | Improved |
| add further alignment step (A-R to A-R-A) (ablation) | ALIGNED (MLP) · E. coli | data/knowledge consistency: data 0.3872 -> 0.3812; knowledge 0.3836 -> 0.4025 | Mixed |
| use a random GRN as knowledge base (ablation) | ALIGNED (GNN) · Norman | balanced consistency: balanced 0.4567 -> 0.3921 | Hurt |
| add inconsistency term (ablation) | ALIGNED (GNN) · Norman | knowledge consistency: knowledge 0.3469 -> 0.4172 | Improved |
| add knowledge-usage restriction L_len (ablation) | ALIGNED (GNN) · Norman | balanced consistency: knowledge 0.4442 -> 0.4172; balanced 0.4715 -> 0.4567 | Mixed |
| learned weight vector instead of random (ablation) | ALIGNED (GNN) · Norman | data/balanced consistency: data 0.4074 -> 0.5101; balanced 0.4813 -> 0.4567 | Mixed |
| ALIGNED vs prior methods without refinement | ALIGNED · perturbation benchmarks (Norman, Dixit, Adamson) | balanced consistency | Improved |
Closest papers by meaning
- PerturbReason: A Knowledge-Grounded Benchmark and Framework for Cell-State–Conditioned Mechanistic Reasoning of Perturbation Effects (NeurIPS 2026 · similarity 0.88)
- Contextualizing biological perturbation experiments through language (ICLR 2025 · similarity 0.88)
- PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimation (NeurIPS 2025 · similarity 0.87)
- BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments (ICLR 2025 · similarity 0.86)
- MapPFN: Learning Causal Perturbation Maps in Context (NeurIPS 2026 · similarity 0.86)
- How Post-Training Shapes Biological Reasoning Models (NeurIPS 2026 · similarity 0.86)
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.
