Unbiased integration of single cell transcriptome replicates
Unbiased integration of single cell transcriptome replicates
复制标题
DOI：
10.1101/2021.05.05.442380
复制
发表时间：
2021-05
影响因子：
4.6
通讯作者：
Martin Loza Lopez;Shunsuke Teraguchi;D. Standley;Diego Diez
中科院分区：
文献类型：
--
作者：
Martin Loza Lopez;Shunsuke Teraguchi;D. Standley;Diego Diez
单细胞转录组学方法正在成为主流，通常使用相同的单细胞技术进行重复实验。无偏数据解释需要通过去除批次效应同时保留生物信息来整合这些数据集的方法。在这里，我们介绍Canek为此目的。Canek利用来自相互最近邻的信息来在模糊逻辑框架内将局部线性校正与细胞特异性非线性校正联合收割机组合。使用组合的真实的和合成数据集，我们表明，Canek纠正批量效应，同时引入最少的偏见相比，竞争的方法。Canek的计算效率很高，可以很容易地从重复实验中整合数千个单细胞转录组。
Single cell transcriptomic approaches are becoming mainstream, with replicate experiments commonly performed with the same single cell technology. Methods that enable integration of these datasets by removing batch effects while preserving biological information are required for unbiased data interpretation. Here we introduce Canek for this purpose. Canek leverages information from mutual nearest neighbor to combine local linear corrections with cell-specific non-linear corrections within a fuzzy logic framework. Using a combination of real and synthetic datasets, we show that Canek corrects batch effects while introducing the least amount of bias compared with competing methods. Canek is computationally efficient and can easily integrate thousands of single-cell transcriptomes from replicated experiments.