GLOBE: a contrastive learning-based framework for integrating single-cell transcriptome datasets

Abstract

Integration of single-cell transcriptome datasets from multiple sources plays an important role in investigating complex biological systems. The key to integration of transcriptome datasets is batch effect removal. Recent methods attempt to apply a contrastive learning strategy to correct batch effects. Despite their encouraging performance, the optimal contrastive learning framework for batch effect removal is still under exploration. We develop an improved contrastive learning-based batch correction framework, GLOBE. GLOBE defines adaptive translation transformations for each cell to guarantee the stability of approximating batch effects. To enhance the consistency of representations alignment, GLOBE utilizes a loss function that is both hardness-aware and consistency-aware to learn batch effect-invariant representations. Moreover, GLOBE computes batch-corrected gene matrix in a transparent approach to support diverse downstream analysis. Benchmarking results on a wide spectrum of datasets show that GLOBE outperforms other state-of-the-art methods in terms of robust batch mixing and superior conservation of biological signals. We further apply GLOBE to integrate two developing mouse neocortex datasets and show GLOBE succeeds in removing batch effects while preserving the contiguous structure of cells in raw data. Finally, a comprehensive study is conducted to validate the effectiveness of GLOBE.

scMAE: a masked autoencoder for single-cell RNA-seq clustering

“…The scMAE method is evaluated on a total of 16 real scRNA-seq datasets, each of which contains cells with known labels or validated in previous studies ( Pollen et al 2014 , Macosko et al 2015 , Baron et al 2016 , Shekhar et al 2016 , Tirosh et al 2016 , Bach et al 2017 , Cao et al 2017 , Guo et al 2018 , Hrvatin et al 2018 , Tabula Muris Consortium et al 2018 , Tosches et al 2018 , Wang et al 2018 , Young et al 2018 , Tran et al 2020 ). These datasets have been widely used for evaluating other clustering methods as well ( Ciortan and Defrance 2021 , Ciortan and Defrance 2022 , Han et al 2022 , Wan et al 2022 , Yan et al 2022 ). The characteristics of these datasets, including the biological tissues and organisms they represent, are summarized in Supplementary Table S1 .…”

CAKE: a flexible self-supervised framework for enhancing cell visualization, clustering and rare cell identification

“…Contrastive learning method is a kind of unsupervised or self-supervised representation learning [ 25 , 26 ]. It involves training a neural network to differentiate between positive and negative instances in a given dataset [ 19 ]. Taking MOCO as an example, the model employs the InfoNCE loss [ 27 ] and momentum-based updating mechanism.…”

“…For scziDesk [ 18 ], it aims to learn more clustering-friendly cell representations by adding a similarity constraint term, and replaces the hard clustering by soft K-means clustering. In addition to autoencoder-based approaches, contrastive learning is a more powerful and flexible self-supervised approach to create informative embeddings for downstream analysis task by capturing the similarity and dissimilarity between cells [ 19 ]. Typically, scNAME [ 20 ] utilizes a neighborhood contrastive loss and an ancillary mask estimation loss to reveal the uncorrupted data structure.…”

Accurate cell type annotation for single‐cell chromatin accessibility data via contrastive learning and reference guidance

“…During the training phase, after feature selection and term frequency-inverse document frequency (TF-IDF) transformation (see more details in Section 5), we utilize contrastive learning to learn latent representations of the training set, which have yielded state-of-theart results on various tasks in single-cell data analysis [28][29][30]. Specifically, we utilize the multi-layer perceptron (MLP) module within the contrastive learning framework, as suggested by recent studies [31][32][33].…”