Liu Siyao, Corcoran David L, Garcia-Recio Susana, Marron James S, Perou Charles M
Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC 27599, United States.
Department of Genetics, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, United States.
NAR Genom Bioinform. 2025 Mar 19;7(1):lqaf023. doi: 10.1093/nargab/lqaf023. eCollection 2025 Mar.
While numerous methods have been developed for analyzing scRNA-seq data, benchmarking various methods remains challenging. There is a lack of ground truth datasets for evaluating novel gene selection and/or clustering methods. We propose the use of , a new approach based upon perturbing signals in a real dataset for comparing analysis methods. We demonstrate the effectiveness of crafted experiments for evaluating new univariate distribution-oriented suite of feature selection methods, called GOF. We show GOF selects features that robustly identify crafted features and perform well on real non-crafted data sets. Using varying ways of crafting, we also show the context in which each GOF method performs the best. GOF is implemented as an open-source R package and freely available under GPL-2 license at https://github.com/siyao-liu/GOF. Source code, including all functions for constructing crafted experiments and benchmarking feature selection methods, are publicly available at https://github.com/siyao-liu/CraftedExperiment.
虽然已经开发出了许多用于分析单细胞RNA测序（scRNA-seq）数据的方法，但对各种方法进行基准测试仍然具有挑战性。缺乏用于评估新型基因选择和/或聚类方法的真实数据集。我们建议使用一种基于在真实数据集中扰动信号来比较分析方法的新方法。我们展示了精心设计的实验对于评估名为GOF的面向单变量分布的新特征选择方法套件的有效性。我们表明，GOF选择的特征能够稳健地识别精心设计的特征，并且在真实的非精心设计的数据集中表现良好。通过使用不同的精心设计方式，我们还展示了每种GOF方法表现最佳的背景。GOF作为一个开源R包实现，并根据GPL-2许可在https://github.com/siyao-liu/GOF上免费提供。包括用于构建精心设计的实验和基准测试特征选择方法的所有函数的源代码可在https://github.com/siyao-liu/CraftedExperiment上公开获取。
