{"type":"ScholarlyArticle","title":"sc2Flow: Mitigating Mean Prediction Bias in Single-Cell Perturbation with Dual-Stage Flow Matching","authors":[{"name":"Hanwen Lyu","slug":"hanwen-lyu","given":"Hanwen","family":"Lyu","affiliation":"Hunan University"},{"name":"Jiawei Luo","slug":"jiawei-luo","given":"Jiawei","family":"Luo","affiliation":"Hunan University"}],"abstract":"Predicting single-cell gene responses to chemical perturbations is vital for personalized therapy, yet existing deep learning models face significant hurdles. Standard regression-based approaches suffer from \"mean prediction bias,\" failing to capture cellular heterogeneity, while current Flow Matching (FM) methods struggle with the extreme sparsity of single-cell data and the dimensionality mismatches inherent in transformer-based generation. To address these challenges, we introduce sc2Flow, a dual-stage framework that unifies discrete and continuous flow matching. sc2Flow first predicts the binary mask of expressed genes and subsequently models their quantitative levels using a scalable Transformer. This decoupling effectively resolves dimensionality conflicts and eliminates parameter redundancy. Extensive benchmarks on Sci-Plex3 and other datasets demonstrate that sc2Flow significantly outperforms state-of-the-art baselines on distribution matching, successfully mitigating mean bias to preserve critical biological heterogeneity.","doi":"10.24963/ijcai.2026/757","url":"https://ijcai.org/proceedings/2026/757","pdf":"https://ijcai.org/proceedings/2026/0757.pdf","markdown":"https://ijcai.org/proceedings/2026/757.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/757","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"AI and Health","track":"AI and Health","number":757,"pages":{"first":6807,"last":6814},"keywords":["Genomic data analysis: Genomic data analysis","Drug discovery: Drug discovery"],"videos":[],"references":[{"position":1,"raw":"[Aliper et al., 2016] Alexander Aliper, Sergey Plis, Artem Artemov, Alvaro Ulloa, Polina Mamoshina, and Alex Zhavoronkov. Deep learning applications for predicting pharmacological properties of drugs and drug repurpos- ing using transcriptomic data. Molecular pharmaceutics, 13(7):2524-2530, 2016.","title":"Deep learning applications for predicting pharmacological properties of drugs and drug repurposing using transcriptomic data","authors":"Alexander Aliper, Sergey Plis, Artem Artemov, Alvaro Ulloa, Polina Mamoshina, Alex Zhavoronkov","year":"2016","venue":"Molecular pharmaceutics","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bishop and Nasrabadi, 2006] Christopher M Bishop and Nasser M Nasrabadi. Pattern recognition and machine learning, volume 4. Springer, 2006.","title":"Pattern recognition and machine learning","authors":"M Christopher, Bishop, M Nasser, Nasrabadi","year":"2006","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Cai et al., 2025] Huanqia Cai, Sihan Cao, Ruoyi Du, Peng Gao, Steven Hoi, Zhaohui Hou, Shijie Huang, Dengyang Jiang, Xin Jin, Liangchen Li, et al. Z-image: An efficient image generation foundation model with single-stream diffusion transformer. arXiv preprint arXiv:2511.22699, 2025.","title":"Z-image: An efficient image generation foundation model with single-stream diffusion transformer","authors":"Huanqia Cai, Sihan Cao, Ruoyi Du, Peng Gao, Steven Hoi, Zhaohui Hou, Shijie Huang, Dengyang Jiang, Xin Jin, Liangchen Li","year":"2025","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Carnero, 2006] Amancio Carnero. High throughput screen- ing in drug discovery. Clinical and Translational Oncol- ogy, 8:482-490, 2006.","title":"High throughput screening in drug discovery","authors":"Amancio Carnero","year":"2006","venue":"Clinical and Translational Oncology","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Gat et al., 2024] Itai Gat, Tal Remez, Neta Shaul, Felix Kreuk, Ricky TQ Chen, Gabriel Synnaeve, Yossi Adi, and Yaron Lipman. Discrete flow matching. Advances in Neu- ral Information Processing Systems, 37:133345-133385, 2024.","title":"Discrete flow matching","authors":"Itai Gat, Tal Remez, Neta Shaul, Felix Kreuk, Ricky Tq Chen, Gabriel Synnaeve, Yossi Adi, Yaron Lipman","year":"2024","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Hetzel et al., 2022] Leon Hetzel, Simon Boehm, Niki Kil- bertus, Stephan Günnemann, Fabian Theis, et al. Pre- dicting cellular responses to novel drug perturbations at a single-cell resolution. Advances in Neural Information Processing Systems, 35:26711-26722, 2022.","title":"Predicting cellular responses to novel drug perturbations at a single-cell resolution","authors":"Leon Hetzel, Simon Boehm, Niki Kilbertus, Stephan Günnemann, Fabian Theis","year":"2022","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Kingma et al., 2013] Diederik P Kingma, Max Welling, et al. Auto-encoding variational bayes, 2013.","title":"Auto-encoding variational bayes","authors":"Max Diederik P Kingma, Welling","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Klein et al., 2025] Dominik Klein, Jonas Simon Fleck, Daniil Bobrovskiy, Lea Zimmermann, Sören Becker, Alessandro Palma, Leander Dony, Alejandro Tejada- Lapuerta, Guillaume Huguet, Hsiu-Chuan Lin, et al. Cellflow enables generative single-cell phenotype model- ing with flow matching. bioRxiv, pages 2025-04, 2025.","title":"Cellflow enables generative single-cell phenotype modeling with flow matching","authors":"Dominik Klein, Jonas Simon Fleck, Daniil Bobrovskiy, Lea Zimmermann, Sören Becker, Alessandro Palma, Leander Dony, Alejandro Tejada-Lapuerta, Guillaume Huguet, Hsiu-Chuan Lin","year":"2025","venue":"bioRxiv","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Li and He, 2025] Tianhong Li and Kaiming He. Back to basics: Let denoising generative models denoise. arXiv preprint arXiv:2511.13720, 2025.","title":"Back to basics: Let denoising generative models denoise","authors":"He ; Li, Tianhong Li, Kaiming He","year":"2025","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Lipman et al., 2022] Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. arXiv preprint arXiv:2210.02747, 2022.","title":"Flow matching for generative modeling","authors":"Yaron Lipman, Ricky Tq Chen, Heli Ben-Hamu, Maximilian Nickel, Matt Le","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Lotfollahi et al., 2019] Mohammad Lotfollahi, F Alexander Wolf, and Fabian J Theis. scgen predicts single-cell pertur- bation responses. Nature methods, 16(8):715-721, 2019.","title":"scgen predicts single-cell perturbation responses","authors":"Mohammad Lotfollahi, Alexander Wolf, Fabian J Theis","year":"2019","venue":"Nature methods","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Lotfollahi et al., 2023] Mohammad Lotfollahi, Anna KCel- lOTlimovskaia Susmelj, Carlo De Donno, Leon Hetzel, Yuge Ji, Ignacio L Ibarra, Sanjay R Srivatsan, Mohsen Naghipourfar, Riza M Daza, Beth Martin, et al. Pre- dicting cellular responses to complex perturbations in high-throughput screens. Molecular systems biology, 19(6):e11517, 2023.","title":"Predicting cellular responses to complex perturbations in high-throughput screens","authors":"Mohammad Lotfollahi, Anna Kcel-Lotlimovskaia Susmelj, Carlo De Donno, Leon Hetzel, Yuge Ji, Ignacio L Ibarra, Mohsen Sanjay R Srivatsan, Naghipourfar, M Riza, Beth Daza, Martin","year":"2023","venue":"Molecular systems biology","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[McFarland et al., 2020] James M McFarland, Brenton R Paolella, Allison Warren, Kathryn Geiger-Schuller, Tsukasa Shibue, Michael Rothberg, Olena Kuksenko, William N Colgan, Andrew Jones, Emily Chambers, et al. Multiplexed single-cell transcriptional response profiling to define cancer vulnerabilities and therapeutic mechanism of action. Nature communications, 11(1):4296, 2020.","title":"Multiplexed single-cell transcriptional response profiling to define cancer vulnerabilities and therapeutic mechanism of action","authors":"Brenton R James M Mcfarland, Allison Paolella, Kathryn Warren, Tsukasa Geiger-Schuller, Michael Shibue, Olena Rothberg, Kuksenko, Andrew William N Colgan, Emily Jones, Chambers","year":"2020","venue":"Nature communications","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Menden et al., 2019] Michael P Menden, Dennis Wang, Mike J Mason, Bence Szalai, Krishna C Bulusu, Yuan- fang Guan, Thomas Yu, Jaewoo Kang, Minji Jeon, Russ Wolfinger, et al. Community assessment to advance computational prediction of cancer drug combinations in a pharmacogenomic screen. Nature communications, 10(1):2674, 2019.","title":"Community assessment to advance computational prediction of cancer drug combinations in a pharmacogenomic screen","authors":"Dennis Michael P Menden, Mike J Wang, Bence Mason, Krishna C Szalai, Yuanfang Bulusu, Thomas Guan, Jaewoo Yu, Minji Kang, Russ Jeon, Wolfinger","year":"2019","venue":"Nature communications","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Piran et al., 2024] Zoe Piran, Niv Cohen, Yedid Hoshen, and Mor Nitzan. Disentanglement of single-cell data with biolord. Nature Biotechnology, 42(11):1678-1683, 2024.","title":"Disentanglement of single-cell data with biolord","authors":"Zoe Piran, Niv Cohen, Yedid Hoshen, Mor Nitzan","year":"2024","venue":"Nature Biotechnology","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Qi et al., 2024] Xiaoning Qi, Lianhe Zhao, Chenyu Tian, Yueyue Li, Zhen-Lin Chen, Peipei Huo, Runsheng Chen, Xiaodong Liu, Baoping Wan, Shengyong Yang, et al. Pre- dicting transcriptional responses to novel chemical pertur- bations using deep generative model for drug discovery. Nature Communications, 15(1):1-19, 2024.","title":"Predicting transcriptional responses to novel chemical perturbations using deep generative model for drug discovery","authors":"Xiaoning Qi, Lianhe Zhao, Chenyu Tian, Yueyue Li, Zhen-Lin Chen, Peipei Huo, Runsheng Chen, Xiaodong Liu, Baoping Wan, Shengyong Yang","year":"2024","venue":"Nature Communications","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Srivatsan et al., 2020] Sanjay R Srivatsan, José L McFaline- Figueroa, Vijay Ramani, Lauren Saunders, Junyue Cao, Jonathan Packer, Hannah A Pliner, Dana L Jackson, Riza M Daza, Lena Christiansen, et al. Massively mul- tiplex chemical transcriptomics at single-cell resolution. Science, 367(6473):45-51, 2020.","title":"Massively multiplex chemical transcriptomics at single-cell resolution","authors":"Sanjay R Srivatsan, L José, Vijay Mcfaline-Figueroa, Lauren Ramani, Junyue Saunders, Jonathan Cao, Hannah A Packer, Pliner, Riza M Dana L Jackson, Lena Daza, Christiansen","year":"2020","venue":"Science","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Su et al., 2024] Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, and Yunfeng Liu. Roformer: En- hanced transformer with rotary position embedding. Neu- rocomputing, 568:127063, 2024.","title":"Roformer: Enhanced transformer with rotary position embedding","authors":"Jianlin Su, Murtadha Ahmed, Yu Lu, Shengfeng Pan, Wen Bo, Yunfeng Liu","year":"2024","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Szalai and Veres, 2023] Bence Szalai and Dániel V Veres. Application of perturbation gene expression profiles in drug discovery-from mechanism of action to quantita- tive modelling. Frontiers in Systems Biology, 3:1126044, 2023.","title":"Application of perturbation gene expression profiles in drug discovery-from mechanism of action to quantitative modelling","authors":"Veres Szalai, Bence Szalai, V Dániel, Veres","year":"2023","venue":"Frontiers in Systems Biology","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Vaswani et al., 2017] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017.","title":"Attention is all you need","authors":"Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, Illia Polosukhin","year":"2017","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Wei et al., 2025] Zhiting Wei, Yiheng Wang, Yicheng Gao, Shuguang Wang, Ping Li, Duanmiao Si, Yuli Gao, Siqi Wu, Danlu Li, Kejing Dong, et al. Benchmarking algo- rithms for generalizable single-cell perturbation response prediction. Nature Methods, pages 1-14, 2025.","title":"Benchmarking algorithms for generalizable single-cell perturbation response prediction","authors":"Wei","year":"2025","venue":"Nature Methods","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhao et al., 2021a] Wenting Zhao, Athanassios Dovas, Eleonora Francesca Spinazzi, Hanna Mendes Levitin, Matei Alexandru Banu, Pavan Upadhyayula, Tejaswi Sud- hakar, Tamara Marie, Marc L Otten, Michael B Sisti, et al. Deconvolution of cell type-specific drug responses in human tumor tissue with single-cell rna-seq. Genome medicine, 13(1):82, 2021.","title":"Deconvolution of cell type-specific drug responses in human tumor tissue with single-cell rna-seq","authors":"Zhao","year":"2021","venue":"Genome medicine","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhao et al., 2021b] Yucheng Zhao, Guangting Wang, Chuanxin Tang, Chong Luo, Wenjun Zeng, and Zheng- Jun Zha. A battle of network structures: An empirical study of cnn, transformer, and mlp. arXiv preprint arXiv:2108.13002, 2021.","title":"A battle of network structures: An empirical study of cnn, transformer, and mlp","authors":"Zhao","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}