2021
DOI: 10.1186/s12859-021-04028-4
|View full text |Cite
|
Sign up to set email alerts
|

scConsensus: combining supervised and unsupervised clustering for cell type identification in single-cell RNA sequencing data

Abstract: Background Clustering is a crucial step in the analysis of single-cell data. Clusters identified in an unsupervised manner are typically annotated to cell types based on differentially expressed genes. In contrast, supervised methods use a reference panel of labelled transcriptomes to guide both clustering and cell type identification. Supervised and unsupervised clustering approaches have their distinct advantages and limitations. Therefore, they can lead to different but often complementary c… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
11
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
7
2
1

Relationship

0
10

Authors

Journals

citations
Cited by 20 publications
(11 citation statements)
references
References 21 publications
0
11
0
Order By: Relevance
“…A key step in the analysis of scRNA-seq data and, more in general, of single cell data, is the identification of cell populations, that is groups of cells sharing similar properties. Several approaches have been proposed to achieve this task, based on well established clustering techniques [ 12 , 13 ], consensus clustering [ 14 16 ] and deep learning [ 17 ]; many more have been recently reviewed [ 18 , 19 ] and benchmarked [ 20 ]. As the popularity of single cell analysis frameworks Seurat [ 21 ] and scanpy [ 22 ] raised, methods based instead on graph partitioning became the de facto standards.…”
Section: Introductionmentioning
confidence: 99%
“…A key step in the analysis of scRNA-seq data and, more in general, of single cell data, is the identification of cell populations, that is groups of cells sharing similar properties. Several approaches have been proposed to achieve this task, based on well established clustering techniques [ 12 , 13 ], consensus clustering [ 14 16 ] and deep learning [ 17 ]; many more have been recently reviewed [ 18 , 19 ] and benchmarked [ 20 ]. As the popularity of single cell analysis frameworks Seurat [ 21 ] and scanpy [ 22 ] raised, methods based instead on graph partitioning became the de facto standards.…”
Section: Introductionmentioning
confidence: 99%
“…Each cluster was attributed with "unique markers", which are the genes expressed only in that type of cell among all the cells sampled, as well as with "combinatorial markers", which are differentially expressed genes that are not restricted to a single cell type. A consensus approach was proposed for both the clustering paradigms in order to increase the accuracy of the clustering and the precision of cell type annotation [72], which is expected to be applied in future studies.…”
Section: Single-cell Transcriptomics Of the Human Telencephalonmentioning
confidence: 99%
“…A problem in single-cell data analysis is cell annotation—assignment of a particular cell type or a cell state to each sequenced cell. The size of the generated datasets made manual annotation approaches utterly unfeasible, while the peculiarities of data generation prompted the development of novel innovative classification methods [ 8 13 ]. This is especially apparent in datasets stemming from cancer tissues, where the variability in the transcriptomic states does not conform to classically defined cell types.…”
Section: Introductionmentioning
confidence: 99%