Single-cell perturbation response models are coupled to regulatory, pathway and temporal…

Preprints

• Co$^2$ODE: Conditionally Coarse-Grained Model Discovery for Rewiring in Biochemical Networks

  arXiv2026-10-30Theodore de Pomereu et al.

  Co$^2$ODE learns sparse neural or symbolic ODEs with a conditional mask and VAE closure for context-dependent rewiring in biochemical networks, applied to…

• CytoWave: Perturbation-Centric Pretraining for Single-Cell Response Prediction

  arXiv2026-11-02Yuwei Miao et al.

  CytoWave: a two-stage perturbation-centric pretraining framework for predicting single-cell gene expression responses to genetic perturbations across datasets…

• Cross-modal transfer enables zero-shot prediction of drug responses across cell types

  bioRxiv2026-11-01Luo, C. et al.

  GRAFT transfers genetic perturbation response information to chemical perturbations for zero-shot prediction of drug-induced transcriptional responses across…

• FateDriver: Weakly supervised dual-view graph learning for transition-specific prioritization of candidate regulators in cellular reprogramming

  bioRxiv2026-11-01Sung, J. et al.

  FateDriver: a weakly supervised dual-view graph neural network that ranks candidate regulators of specified cell-state transitions in reprogramming, using…

• EMBER: Contextualized gene representations enable compositional prediction across cellular perturbations

  bioRxiv2026-10-31Ho, N. et al.

  EMBER, a gene-centric foundation model with frozen contextualized gene embeddings and lightweight probes, predicts single, digenic, and drug-response…

• GSET: Gene Signals from Embedding Trajectories select compact contexts for mechanistic GRN inference

  bioRxiv2026-11-03WEIZMANN, Z. et al.

  GSET uses optimal transport residuals and KL divergence on temporal gene embeddings to select compact, dynamically informed gene sets for mechanistic GRN…

• RELAY: Relational In-Context Learning for Cellular Perturbation Prediction

  bioRxiv2026-11-03Sun, Y. et al.

  RELAY is a relational in-context model that predicts unmeasured cellular perturbation responses from measured control-perturbed pairs sharing experimental…

• ExPert infers causal perturbations from cell states across biological and clinical contexts

  bioRxiv2026-11-03Schwartz, L. et al.

  ExPert learns perturbation prototypes in a shared latent space across 12 cellular contexts to infer causal perturbations from observed cell states, with…

• Spectral Adaptive Repositioning for Flow-Based Single-Cell Perturbation Modeling

  bioRxiv2026-11-04Verma, S. et al.

  SPEAR is a flow-based generative model for single-cell perturbation response prediction, using spectral Laplacian gene encodings and perturbation-conditioned…

• Resolving how genetic perturbations act through gene regulatory paths

  bioRxiv2026-11-03Ren, L. et al.

  REGA-perturb infers how genetic perturbations act via TF-regulatory element-gene paths, using reference epigenomes and pooled latent programs from Perturb-seq…

• RAPTORGraph: graph-based pathway modeling for causal discovery in single-cell perturbations

  bioRxiv2026-11-04Adhisantoso, Y. G. et al.

  RAPTORGraph is a beta-VAE with a graph pathway encoder for causal discovery in Perturb-seq single-cell perturbation data, aiming to eliminate intervention…

Related topics

• Biologically Constrained Models For Omics
• Single Cell Perturbation Response Modeling
• Repurposing Single Cell Foundation Models

Topic history