Many perturbation models are coming out!
| Name | Year | Journal/Conf. | Title |
|---|---|---|---|
| Rachel et al | 2018 | Pacific Symposium on Biocomputing 2018 | Cell-specific prediction and application of drug-induced gene expression profiles |
| scGEN | 2019 | Nature Method | scGen predicts single-cell perturbation responses |
| DTD | 2019 | The World Wide Web Conference, 2019 | Modeling Relational Drug-Target-Disease Interactions via Tensor Factorization with Multiple Web Sources |
| CPA | 2021 | Molecular system biology | Predicting cellular responses to complex perturbations in high‐throughput screens |
| CellBox | 2021 | Cell systems | CellBox: Interpretable Machine Learning for Perturbation Biology with Application to the Design of Cancer Combination Therapy |
| CellDrift | 2022 | BIB | CellDrift: inferring perturbation responses in temporally sampled single-cell data |
| MultiCPA | 2022 | MultiCPA: Multimodal Compositional Perturbation Autoencoder | |
| PerturbNet | 2022 | PerturbNet predicts single-cell responses to unseen chemical and genetic perturbations | |
| scINSIGHT | 2022 | Genome biology | scINSIGHT for interpreting single-cell gene expression from biologically heterogeneous data |
| scpregan | 2022 | Bioinformatics | scPreGAN, a deep generative model for predicting the response of single-cell expression to perturbation |
| Gears | 2023 | Nature Biotech | Predicting transcriptional outcomes of novel multigene perturbations with GEARS |
| cycleCDR | 2023 | Interpretable Modeling of Single-cell perturbation Responses to Novel Drugs Using Cycle Consistence Learning | |
| scVIDR | 2023 | Patterns | Generative modeling of single-cell gene expression for dose-dependent chemical perturbations |
| Unagi | 2023 | Unagi: Deep Generative Model for Deciphering Cellular Dynamics and In-Silico Drug Discovery in Complex Diseases | |
| CINEMA-OT | 2023 | Nature Method | Causal identification of single-cell experimental perturbation effects with CINEMA-OT |
| ChemCPA | 2023 | NeurIPS 2022 | Predicting Cellular Responses to Novel Drug Perturbations at a Single-Cell Resolution |
| DREEP | 2023 | BMC Medicine | Predicting drug response from single-cell expression profiles of tumours |
| ontoVAE | 2023 | Bioinformatics | Biologically informed variational autoencoders allow predictive modeling of genetic and drug-induced perturbations |
| scDiff | 2023 | A GENERAL SINGLE-CELL ANALYSIS FRAMEWORK VIA CONDITIONAL DIFFUSION GENERATIVE MODELS | |
| ContrastiveVI | 2023 | Nature Method | Isolating salient variations of interest in single-cell data with contrastiveVI |
| sVAE | 2023 | PMLR | Learning Causal Representations of Single Cells via Sparse Mechanism Shift Modeling |
| CellOT | 2023 | Nature Method | Learning single-cell perturbation responses using neural optimal transport |
| MOASL | 2023 | Computers in Biology and Medicine | MOASL: Predicting drug mechanism of actions through similarity learning with transcriptomic signature |
| samsVAE | 2024 | Advances in Neural Information Processing Systems | Modelling Cellular Perturbations with the Sparse Additive Mechanism Shift Variational Autoencoder |
| Biolord | 2024 | Nature Biotech | Disentanglement of single-cell data with biolord |
| PDGrapher | 2024 | Nature Biomedical Engineering | Combinatorial prediction of therapeutic perturbations using causally-inspired neural networks |
| TAT | 2024 | Journal of Chemical Information and Modeling | Compound Activity Prediction with Dose-Dependent Transcriptomic Profiles and Deep Learning |
| scVAE | 2024 | A Supervised Contrastive Framework for Learning Disentangled Representations of Cell Perturbation Data | |
| Cell PaintingCNN | 2024 | NC | Learning representations for image-based profiling of perturbations |
| scDisInFact | 2024 | NC | scDisInFact: disentangled learning for integration and prediction of multi-batch multi-condition single-cell RNA-sequencing data |
| CellCap | 2024 | Modeling interpretable correspondence between cell state and perturbation response with CellCap | |
| CODEX | 2024 | Bioinformatics | CODEX: COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations |
| scFM | 2024 | PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction | |
| STAMP | 2024 | NCS | Toward subtask-decomposition-based learning and benchmarking for predicting genetic perturbation outcomes and beyond |
| PrePR-CT | 2024 | PrePR-CT: Predicting Perturbation Responses in Unseen Cell Types Using Cell-Type-Specific Graphs | |
| PRnet | 2024 | NC | Predicting transcriptional responses to novel chemical perturbations using deep generative model for drug discovery |
| TranSiGen | 2024 | NC | Deep representation learning of chemical-induced transcriptional profile for phenotype-based drug discovery |
| BioDiscoveryAgent | 2024 | BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments | |
| DRSPRING | 2024 | Computers in Biology and Medicine | DRSPRING: Graph convolutional network (GCN)-Based drug synergy prediction utilizing drug-induced gene expression profile |
| PertKGE | 2024 | Identify compound-protein interaction with knowledge graph embedding of perturbation transcriptomics | |
| scRank | 2024 | Cell Reports Medicine | scRank infers drug-responsive cell types from untreated scRNA-seq data using a target-perturbed gene regulatory network |
| CellFlow | 2025 | CellFlow enables generative single-cell phenotype modeling with flow matching | |
| TxPert | 2025 | TxPert: Leveraging Biochemical Relationships for Out-of-Distribution Transcriptomic Perturbation Prediction | |
| IMPA | 2025 | NC | Predicting cell morphological responses to perturbations using generative modeling |
| PS | 2025 | Nature Cell Biology | Decoding heterogeneous single-cell perturbation responses |
| TRADE | 2025 | Nature Genetics | Transcriptome-wide analysis of differential expression in perturbation atlases |
| UNAGI | 2025 | Nature Biomedical Engineering | A deep generative model for deciphering cellular dynamics and in silico drug discovery in complex diseases |
| STATE | 2025 | Predicting cellular responses to perturbation across diverse contexts with State | |
| scAgent | 2025 | scAgents:AMulti-AgentFramework forFullyAutonomousEnd-to-EndSingle-Cell PerturbationAnalysis | |
| CONCERT | 2025 | CONCERT predicts niche-aware perturbation responses in spatial transcriptomics | |
| Cradle-VAE | 2025 | AAAI Conference on AI | Cradle-VAE: Enhancing Single-Cell Gene Perturbation Modeling with Counterfactual Reasoning-based Artifact Disentanglement |
| Pertpy | 2025 | NM | Pertpy: an end-to-end framework for perturbation analysis |
| Scouter | 2025 | NCS | Scouter predicts transcriptional responses to genetic perturbations with large language model embeddings |
| PertAdapt | 2025 | PertAdapt: Unlocking Single-Cell Foundation Models for Genetic Perturbation Prediction via Condition-Sensitive Adaptation | |
| CausalGRN | 2025 | CausalGRN: deciphering causal gene regulatory networks from single-cell CRISPR screens | |
| Systema | 2025 | Nature Biotechnology | Systema: a framework for evaluating genetic perturbation response prediction beyond systematic variation |
| Stack | 2026 | Stack: In-Context Learning of Single-Cell Biology | |
| AlphaCell | 2026 | Towards building a World Model to simulate perturbation-induced cellular dynamics by AlphaCell | |
| XPert | 2026 | NMI | Modelling drug-induced cellular perturbation responses with a biologically informed dual-branch transformer |
| PrePR-CT | 2026 | NMI | Predicting and interpreting cell-type-specific drug responses in the small-data regime using inductive priors |
| SequenTx | 2026 | NMI | Reinforcement learning-based design of sequential drug treatment targeting the evolving tumour landscape with SequenTx |
|2026|Nature Reviews Genetics| Interpretation, extrapolation and perturbation of single cells
|2025|BMC genomics| Benchmarking foundation cell models for post-perturbation RNA-seq prediction
|2025| NM | Deep-learning-based gene perturbation effect prediction does not yet outperform simple linear baselines
|2025| Communications Biology | A large-scale benchmark for network inference from single-cell perturbation data
|2025| Bioinformatics | Simple controls exceed best deep learning algorithms and reveal foundation model effectiveness for predicting genetic perturbations
SC-perturb
C-MAP
PerturbBase
PerturDB
Tahoe-100M
Multiome Perturb-seq (paper)
Multiome: CRISPRmap (paper)
Spatial: Perturb-Fish (paper)
Spatial: PerturbView (paper)
Spatial: Perturb-map (paper)
Spatial: Perturb-DBiT (paper)
Spatial: Spatial-perturb-seq (paper)
Spatial: NIS-seq (paper)
Spatial: PERTURB-CAST (paper)
List of scFM
We will build APIs for some of these models on TDC for benchmarking.
Find single-cell data from CZI database
A tool to quickly check:
Please contact Xiang Lin (xiang_lin@hms.harvard.edu).
@misc{pertogether2025,
title = {PerTogether: A tool for perturbation research and modeling},
author = {Xiang Lin},
year = {2025},
url = {https://github.com/xianglin226/Benchmarking-Single-Cell-Perturbation/},
note = {Version 1.0}
}