scAMACE: model-based approach to the joint analysis of single-cell data on chromatin accessibility, gene expression and methylation

Abstract

Motivation The advancement in technologies and the growth of available single-cell datasets motivate integrative analysis of multiple single-cell genomic datasets. Integrative analysis of multimodal single-cell datasets combines complementary information offered by single-omic datasets and can offer deeper insights on complex biological process. Clustering methods that identify the unknown cell types are among the first few steps in the analysis of single-cell datasets, and they are important for downstream analysis built upon the identified cell types. Results We propose scAMACE for the integrative analysis and clustering of single-cell data on chromatin accessibility, gene expression and methylation. We demonstrate that cell types are better identified and characterized through analyzing the three data types jointly. We develop an efficient expectationmaximization (EM) algorithm to perform statistical inference, and evaluate our methods on both simulation study and real data applications. We also provide the GPU implementation of scAMACE, making it scalable to large datasets. Availability The software and datasets are available at https://github.com/cuhklinlab/scAMACE_py (python implementation) and https://github.com/cuhklinlab/scAMACE (R implementation). Supplementary information Supplementary data are available at Bioinformatics online.

iPoLNG—An unsupervised model for the integrative analysis of single-cell multiomics data

“…Integration problems in single-cell biology can be divided into those associated with the integration of unmatched data (that is, different modalities profiled from different cells) or matched (that is, different modalities profiled from the same cell) data ( Miao et al, 2021 ). A few methods have been developed for the integrative analysis of unmatched data ( Duren et al, 2018 ; Zeng et al, 2019 ; Cao et al, 2020 ; Lin et al, 2020 ; Wangwu et al, 2021 ; Zeng et al, 2021 ; Zeng and Lin, 2021 ; Cao et al, 2022 ; Demetci et al, 2022 ), which are not applicable to matched data. Some matched data integration methods ( Kim et al, 2020 ; Wang et al, 2020 ; Gayoso et al, 2021 ) are designed for technologies that jointly profile transcriptomic and surface protein data such as CITE-seq ( Stoeckius et al, 2017 ) and REAP-seq ( Peterson et al, 2017 ).…”

“…modalities profiled from the same cell) data (Miao et al, 2021). A few methods have been developed for the integrative analysis of unmatched data (Duren et al, 2018;Zeng et al, 2019;Cao et al, 2020;Lin et al, 2020;Wangwu et al, 2021;Cao et al, 2022;Demetci et al, 2022), which are not applicable to matched data. Some matched data integration methods (Kim et al, 2020;Wang et al, 2020;Gayoso et al, 2021) are designed for technologies that jointly profile transcriptomic and surface protein data such as CITE-seq (Stoeckius et al, 2017) and REAPseq (Peterson et al, 2017).…”

Scalable Integration of Multiomic Single Cell Data Using Generative Adversarial Networks

“…To the best of our knowledge, a fully integrated and generalizable way to analyse such data is still missing. The main solutions proposed so far refer to Manifold Alignment (MA) applications and Deep Learning (DL) (28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42). Both approaches share the final goal of representing multiple feature sets in a common manifold embedding.…”

scICML: Information-theoretic Co-clustering-based Multi-view Learning for the Integrative Analysis of Single-cell Multi-omics data

“…The methods of data integration are growing recently, and most of them are designed for data measured in different cells and sampled from the same cell population. These methods include Seurat (versions 2 and 3) [4,25], MOFA [3], coupleNMF [17], scVDMC [32], LIGER [28], scACE [21], coupleCoC [30], coupleCoC+ [31], scMC [33] and scAMACE [27]. A more comprehensive discussion on integration of single-cell genomic data is presented in [15].…”