arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.04901q-bio.GNcs.LG
收藏

超越独立基因:学习模块归纳表示用于单细胞基因扰动预测

Beyond Independent Genes: Learning Module-Inductive Representations for Single-Cell Gene Perturbation Prediction

The State Key Lab of Brain-Machine Intelligence, Zhejiang University, Hangzhou, Zhejiang, China(脑机智能国家重点实验室,浙江大学,杭州,浙江省,中国);ReLER Lab, College of Artificial Intelligence, Zhejiang University, Hangzhou, Zhejiang, China(ReLER实验室,人工智能学院,浙江大学,杭州,浙江省,中国);Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA(生物医学信息学系,哈佛医学院,波士顿,马萨诸塞州,美国)

Jiafa Ruan, Ruijie Quan, Liyang Xu, Zongxin Yang, Yi Yang

AI总结:

提出scBIG框架,通过基因关系聚类、基因簇感知编码器和结构感知对齐学习协调的基因程序模块表示,结合条件流匹配实现灵活泛化的扰动预测,在多个单细胞扰动基准上平均提升6.7%。

摘要:

预测遗传扰动引起的转录响应是功能基因组学中的一个核心问题。实际上,扰动响应很少是基因独立的,而是表现为功能相关基因之间协调的、程序级别的转录变化。然而,大多数现有方法由于基于基因的建模范式以及依赖无法捕捉动态程序重组的静态生物学先验知识,未能显式建模这种协调性。为解决这些局限,我们提出scBIG,一种模块归纳的扰动预测框架,显式建模协调的基因程序。scBIG通过基因关系聚类从数据中归纳出连贯的基因程序,通过基因簇感知编码器捕获程序间交互,并使用结构感知对齐目标保持模块协调性。然后利用条件流匹配对这些结构化表示进行建模,以实现灵活且可泛化的扰动预测。在多个单细胞扰动基准上的大量实验表明,scBIG始终优于最先进的方法,特别是在未见和组合扰动设置中,相比最强基线平均提升6.7%。代码可在该https URL获取。

英文摘要:

Predicting transcriptional responses to genetic perturbations is a central problem in functional genomics. In practice, perturbation responses are rarely gene-independent but instead manifest as coordinated, program-level transcriptional changes among functionally related genes. However, most existing methods do not explicitly model such coordination, due to gene-wise modeling paradigms and reliance on static biological priors that cannot capture dynamic program reorganization. To address these limitations, we propose scBIG, a module-inductive perturbation prediction framework that explicitly models coordinated gene programs. scBIG induces coherent gene programs from data via Gene-Relation Clustering, captures inter-program interactions through a Gene-Cluster-Aware Encoder, and preserves modular coordination using structure-aware alignment objectives. These structured representations are then modeled using conditional flow matching to enable flexible and generalizable perturbation prediction. Extensive experiments on multiple single-cell perturbation benchmarks show that scBIG consistently outperforms state-of-the-art methods, particularly on unseen and combinatorial perturbation settings, achieving an average improvement of 6.7% over the strongest baselines. The code is available at https://github.com/ttruan2426-dot/scBIG.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑