Table 1.

Methods for analysis sc-RNA sequencing data in the breast cancer.

Algorithm name	Function	Ref.
Nanogrid single-nucleus RNA sequencing	Developed a high-throughput 3′single-core RNA sequencing method, which combines nano-grid technology, automatic imaging, and cell selection, and can sequence up to 1800 single-cores in parallel.	55
ISOP method	It provides a novel method to express the isoform level and heterogeneity in single-cell RNA sequencing data.	56
UQ-pgQ2 combined with DESeq2	It improves the analysis based on intra-group comparison and applies it to the public RNA-seq breast cancer data set.	57
Average-based approach (gene-level expression) to isoform abundance/splicing event	It highlights the importance of splicing mechanisms in defining tumor heterogeneity.	58
SSCA and SSCVA methods	It can recover known biological characteristics from the data set and the shallow sparse connection autoencoders used for gene set projection.	59
SCmutt	It is a new and reliable statistical method which identifies specific cells with mutations found in bulk cell data.	60
CSMF method	It can reveal common and specific pattern scenarios with important biological significance from interrelated biological data.	61
EVA	It is used for evaluating the heterogeneity of gene expression in pathways or gene sets in single-cell RNA-seq data.	62
Digitaldlsorter	The algorithm deep learning scRNA-Seq deconvolution gene expression data.	63
VDJView	It can mine and analyze single-cell multi-omics data.	64
DUSC	The system integrates feature generation based on deep learning architecture and model-based clustering algorithms to obtain compact and useful single-cell transcription data.	65

CSMF common and specific patterns via matrix factorization, DUSC deep unsupervised single-cell clustering, ISOP ISOform-patterns, EVA expression variation analysis, SSCAs shallow sparsely connected autoencoders, SSCVAs shallow sparsely connected variational autoencoders.