A completely parameter-free method for graph-based single cell RNA-seq clustering

Zand, Maryam; Ruan, Jianhua

doi:10.1101/2021.07.15.452521

Cited by 2 publications

(3 citation statements)

References 42 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the clustering performance, we compared ADClust with other tools including MultiK, SIMLR, scDeepCluster, SC3, scQcut [36], IKAP, CIDR [37], Seurat (version 3.0), and DESC. As MultiK outputs multiple estimated cluster number, we selected the estimated cluster number with the highest ARI.…”

Section: Benchmark Methodsmentioning

confidence: 99%

“…As MultiK outputs multiple estimated cluster number, we selected the estimated cluster number with the highest ARI. We set the "NUMC" parameter of SIMLR as a range [2:20] to estimate the cluster number followed ref [36]. We set the true cluster numbers for scDeepCluster since it could not estimate the cluster numbers.…”

Section: Benchmark Methodsmentioning

confidence: 99%

“…This value was 10% higher than Seurat, another graph partitioning algorithm. The better performance by scQcut is likely because scQcut optimized the number of neighbors for the KNN-graph [36]. The 4 rd ranked method DESC achieved decent performance since it jointly optimized cell labels assignment and learned the latent representation that was fitted for clustering.…”

Section: Performance On Scrna-seq Datasetsmentioning

confidence: 99%

See 2 more Smart Citations

A parameter-free deep embedded clustering method for single-cell RNA-seq data

Zeng

Wei

Zhong

et al. 2022

Briefings in Bioinformatics

View full text Add to dashboard Cite

Clustering analysis is widely used in single-cell ribonucleic acid (RNA)-sequencing (scRNA-seq) data to discover cell heterogeneity and cell states. While many clustering methods have been developed for scRNA-seq analysis, most of these methods require to provide the number of clusters. However, it is not easy to know the exact number of cell types in advance, and experienced determination is not always reliable. Here, we have developed ADClust, an automatic deep embedding clustering method for scRNA-seq data, which can accurately cluster cells without requiring a predefined number of clusters. Specifically, ADClust first obtains low-dimensional representation through pre-trained autoencoder and uses the representations to cluster cells into initial micro-clusters. The clusters are then compared in between by a statistical test, and similar micro-clusters are merged into larger clusters. According to the clustering, cell representations are updated so that each cell will be pulled toward centers of its assigned cluster and similar clusters, while cells are separated to keep distances between clusters. This is accomplished through jointly optimizing the carefully designed clustering and autoencoder loss functions. This merging process continues until convergence. ADClust was tested on 11 real scRNA-seq datasets and was shown to outperform existing methods in terms of both clustering performance and the accuracy on the number of the determined clusters. More importantly, our model provides high speed and scalability for large datasets.

show abstract

Section: Benchmark Methodsmentioning

confidence: 99%

Section: Benchmark Methodsmentioning

confidence: 99%

Section: Performance On Scrna-seq Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

A parameter-free deep embedded clustering method for single-cell RNA-seq data

Zeng

Wei

Zhong

et al. 2022

Briefings in Bioinformatics

View full text Add to dashboard Cite

show abstract

A Parameter-free Deep Embedded Clustering Method for Single-cell RNA-seq Data

Zeng

Wei

Zhong

et al. 2021

Preprint

View full text Add to dashboard Cite

Clustering analysis is widely utilized in single-cell RNA-sequencing (scRNA-seq) data to discover cell heterogeneity and cell states. While many clustering methods have been developed for scRNA-seq analysis, most of these methods require to provide the number of clusters. However, it is not easy to know the exact number of cell types in advance, and experienced determination is not always reliable. Here, we have developed ADClust, an automatic deep embedding clustering method for scRNA-seq data, which can accurately cluster cells without requiring a predefined number of clusters. Specifically, ADClust first obtains low-dimensional representation through pre-trained autoencoder, and uses the representations to cluster cells into initial micro-clusters. The clusters are then compared in between by a statistical test, and similar micro-clusters are merged into larger clusters. According to the clustering, cell representations are updated so that each cell will be pulled toward centres of its assigned cluster and similar clusters, while cells are separated to keep distances between clusters. This is accomplished through jointly optimizing the carefully designed clustering and autoencoder loss functions. This merging process continues until convergence. ADClust was tested on eleven real scRNA-seq datasets, and shown to outperform existing methods in terms of both clustering performance and the accuracy on the number of the determined clusters. More importantly, our model provides high speed and scalability for large datasets.

show abstract

A completely parameter-free method for graph-based single cell RNA-seq clustering

Cited by 2 publications

References 42 publications

A parameter-free deep embedded clustering method for single-cell RNA-seq data

A parameter-free deep embedded clustering method for single-cell RNA-seq data

A Parameter-free Deep Embedded Clustering Method for Single-cell RNA-seq Data

Contact Info

Product

Resources

About