Rachel (Menghua) Wu
@rachelwu.bsky.social
phd student at mit. machine learning for graphs, molecules, and biology.
We find
🧬 Existing methods perform poorly on PerturbQA ❌
🧬 Naively applying LLMs also performs poorly ❌
🧬 Simple reasoning template + retrieving experimental outcomes = does ok ✅
🧬 LLMs easily summarize gene sets ✅
🧬 Still a long way to go! ✨
🧬 Existing methods perform poorly on PerturbQA ❌
🧬 Naively applying LLMs also performs poorly ❌
🧬 Simple reasoning template + retrieving experimental outcomes = does ok ✅
🧬 LLMs easily summarize gene sets ✅
🧬 Still a long way to go! ✨
March 4, 2025 at 12:22 AM
We find
🧬 Existing methods perform poorly on PerturbQA ❌
🧬 Naively applying LLMs also performs poorly ❌
🧬 Simple reasoning template + retrieving experimental outcomes = does ok ✅
🧬 LLMs easily summarize gene sets ✅
🧬 Still a long way to go! ✨
🧬 Existing methods perform poorly on PerturbQA ❌
🧬 Naively applying LLMs also performs poorly ❌
🧬 Simple reasoning template + retrieving experimental outcomes = does ok ✅
🧬 LLMs easily summarize gene sets ✅
🧬 Still a long way to go! ✨
🧬 Existing methods use knowledge graphs to relate seen vs. unseen perturbations. It seems wrong that current GNNs discard the semantics of knowledge graphs
🧬 Language may be useful for harmonizing these diverse information
🧬 Language may be useful for harmonizing these diverse information
March 4, 2025 at 12:22 AM
🧬 Existing methods use knowledge graphs to relate seen vs. unseen perturbations. It seems wrong that current GNNs discard the semantics of knowledge graphs
🧬 Language may be useful for harmonizing these diverse information
🧬 Language may be useful for harmonizing these diverse information
We propose
🧬 PerturbQA: a new benchmark for perturbations + LLM reasoning for biomolecular discovery.
🧬 Perturbation experiments should be modeled at the granularity of statistical insights. Goal = predict differential expression / characterize gene sets
🧬 PerturbQA: a new benchmark for perturbations + LLM reasoning for biomolecular discovery.
🧬 Perturbation experiments should be modeled at the granularity of statistical insights. Goal = predict differential expression / characterize gene sets
March 4, 2025 at 12:22 AM
We propose
🧬 PerturbQA: a new benchmark for perturbations + LLM reasoning for biomolecular discovery.
🧬 Perturbation experiments should be modeled at the granularity of statistical insights. Goal = predict differential expression / characterize gene sets
🧬 PerturbQA: a new benchmark for perturbations + LLM reasoning for biomolecular discovery.
🧬 Perturbation experiments should be modeled at the granularity of statistical insights. Goal = predict differential expression / characterize gene sets