Removing batch effects in scRNA-seq data

To fully utilize the power of single-cell RNA sequencing (scRNA-seq) technologies for identifying cell lineages and bona fide transcriptional signals, it is necessary to combine data from multiple experiments. A team of researchers from Indiana University Bloomington and the Indiana University School of Medicine have developed BERMUDA (Batch Effect ReMoval Using Deep Autoencoders), a novel transfer-learning-based method for batch effect correction in scRNA-seq data. BERMUDA effectively combines different batches of scRNA-seq data with vastly different cell population compositions and amplifies biological signals by transferring information among batches. The researchers demonstrate that BERMUDA outperforms existing methods for removing batch effects and distinguishing cell types in multiple simulated and real scRNA-seq datasets.

Overview of BERMUDA for removing batch effects in scRNA-seq data

rna-seq

a The workflow of BERMUDA. Circles and triangles represent cells from Batch 1 and Batch 2, respectively. Different colors represent different cell types. A graph-based clustering algorithm was first applied on each batch individually to detect cell clusters. Then, MetaNeighbor, a method based on Spearman correlation, was used to identify similar clusters between batches. An autoencoder was subsequently trained to perform batch correction on the code of the autoencoder. The code of the autoencoder is a low-dimensional representation of the original data without batch effects and can be used for further analysis. bTraining an autoencoder to remove batch effects. The blue solid lines represent training with the cells in Batch 1 and the blue dashed lines represent training with cells in Batch 2. The black dashed lines represent the calculation of losses. The loss function we optimized contains two components: the reconstruction loss between the input and the output of the autoencoder, and the MMD-based transfer loss between the codes of similar clusters

Availability – The implementation of BERMUDA can be downloaded from Github (https://github.com/txWang/BERMUDA)

Wang T, Johnson TS, Shao W, Lu Z, Helm BR, Zhang J, Huang K. (2019) BERMUDA: a novel deep transfer learning method for single-cell RNA sequencing batch correction reveals hidden high-resolution cellular subtypes. Genome Biol [Epub ahead of print]. [article]

Leave a Reply

Your email address will not be published. Required fields are marked *

*

Time limit is exhausted. Please reload CAPTCHA.