Batch effects in single-cell RNA sequencing data are corrected by matching mutual nearest neighbours

By Laleh Haghverdi, Aaron T. L. Lun, Martin T. Morgan, John C. Marioni. Nature Biotechnology, 2018

2,133 citations recorded by the source corpus.

Large-scale single-cell RNA sequencing (scRNA-seq) data sets that are produced in different laboratories and at different times contain batch effects that may compromise the integration and interpretation of the data. Existing scRNA-seq analysis methods incorrectly assume that the composition of cell populations is either known or identical across batches. We present a strategy for batch correction based on the detection of mutual nearest neighbors (MNNs) in the high-dimensional expression space. Our approach does not rely on predefined or equal population compositions across batches; instead, it requires only that a subset of the population be shared between batches. We demonstrate the superiority of our approach compared with existing methods by using both simulated and real scRNA-seq data sets. Using multiple droplet-based scRNA-seq data sets, we demonstrate that our MNN batch-effect-correction method can be scaled to large numbers of cells.

Subfields: Cancer Research, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/56bf103a956b4a2ad11868c05db189d1e3f64ac1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cancer Research

In Molecular Biology

Citations in the corpus

Fully-automated and ultra-fast cell-type identification using specific marker combinations from single-cell transcriptomic data (2022, background): “However, such supervised approaches require that the reference and new datasets resemble each other, which often pose a problem in scRNA-seq studies 16 .”
MetaCell: analysis of single-cell RNA-seq data using K-nn graph partitions (2019, methodology): “The approach is somewhat similar to methods using mutual K -nn analysis to normalize batch effects [38, 39], or more generally to approaches using symme-trization of the K -nn graph to facilitate dimensionality reduction [40].”
Spatially informed clustering, integration, and deconvolution of spatial transcriptomics with GraphST (2023, positive): “Our GraphST framework removes batch effects implicitly during model training, unlike most existing approaches that remove batch effects by explicitly detecting batch factors, such as Harmony 16 , the mutual nearest neighbors (MNN) approach 35 , and Seurat 3.0 8 .”
Computational principles and challenges in single-cell data integration (2021, methodology): “Instead of integrating all embryos simultaneously, the researchers performed horizontal integration with MNN 21 in a bottom–up fashion, starting with the closely related E6.”
A Single-Cell Atlas of In Vivo Mammalian Chromatin Accessibility. (2018, positive): “Both limitations are likely addressable via optimization and adoption of a mutual nearestneighbors approach (Haghverdi et al., 2018), respectively.”
Efficient integration of heterogeneous single-cell transcriptomes using Scanorama (2019, methodology): “8–9) but now have greater power to detect rare cell populations.”
Structural Remodeling of the Human Colonic Mesenchyme in Inflammatory Bowel Disease (2018, methodology): “For each dataset, we also computed entropy of batch mixing, as described by (Haghverdi et al., 2018), for tSNE cell embeddings of sample batches.”