Reproduce “CytoWave: Perturbation-Centric Pretraining for Single-Cell Response Prediction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuwei Miao, Jingquan Yan, Hehuan Ma, Thao M. Dang, Lin Xu, Siyuan Zhang, Junzhou Huang
CytoWave predicts a cell’s post-perturbation gene expression from its starting expression profile and the genetic perturbation. It first learns reference cell states, then pretrains on perturbation responses using direct prediction and contrastive alignment. On stimulated DatlingerBock2017, CytoWave-S2 scores 0.330 ± 0.002 ΔPCC20 versus 0.300 ± 0.008 for S1, after dataset-specific fine-tuning; this comparison does not isolate alignment’s effect.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.