Chem2Gen-Bench is a new benchmark and evaluation framework for measuring alignment between chemical and genetic perturbation responses in matched cell-target contexts using retrieval, similarity, and embedding comparisons.
Ibarra, Sanjay R
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6verdicts
UNVERDICTED 6representative citing papers
OCOO-T is a flow-matching Transformer model that directly denoises continuous gene expression profiles to predict transcriptional responses to perturbations and reports state-of-the-art results on Tahoe100M, Replogle, and PBMC benchmarks.
Under Bemis-Murcko scaffold split on THP-1 DRUG-seq data, inverse-variance proxy ranks linear Morgan fingerprint regression highest while contest wMSE ranks deep fusion models highest, with fusion beating linear by -0.012 wMSE (p<10^-4).
MEDAL distills manifold embeddings into autoencoders to enable out-of-sample extension and held-out validation of dimension reduction methods.
K-nearest neighbor from a knowledge graph beats most methods on out-of-distribution transcriptomic perturbation prediction, and an RL-trained reasoning LLM matches SOTA on Replogle et al. (2022) cell lines while improving downstream differential expression prediction.
A new harmonized multi-dataset resource of small molecule perturbation transcriptomics is introduced, revealing weak cross-dataset logFC agreement but improved compound embedding performance in held-out evaluations.
citing papers explorer
-
Chem2Gen-Bench: Benchmarking Chemical-to-Genetic Translation in Perturbation Response Space
Chem2Gen-Bench is a new benchmark and evaluation framework for measuring alignment between chemical and genetic perturbation responses in matched cell-target contexts using retrieval, similarity, and embedding comparisons.
-
The Metric Picks the Winner: Evaluation Choice Flips Model Rankings for Drug-Response Prediction in Unseen Chemistry
Under Bemis-Murcko scaffold split on THP-1 DRUG-seq data, inverse-variance proxy ranks linear Morgan fingerprint regression highest while contest wMSE ranks deep fusion models highest, with fusion beating linear by -0.012 wMSE (p<10^-4).
-
Knowledge Graphs and Reasoning LLMs for Finding Simple Yet Effective Transcriptomic Perturbation Predictors
K-nearest neighbor from a knowledge graph beats most methods on out-of-distribution transcriptomic perturbation prediction, and an RL-trained reasoning LLM matches SOTA on Replogle et al. (2022) cell lines while improving downstream differential expression prediction.
-
Chem-PerturBridge: a harmonized compendium of small molecule perturbation transcriptomic effects
A new harmonized multi-dataset resource of small molecule perturbation transcriptomics is introduced, revealing weak cross-dataset logFC agreement but improved compound embedding performance in held-out evaluations.