scCCESS

The software tool 'scCCESS' introduces an autoencoder-based cluster ensemble framework for enhanced cell type identification in single-cell RNA-sequencing (scRNA-seq) data analysis. Addressing the challenges of high feature-dimensionality in transcriptome profiles, the framework employs random subspace projections, autoencoder neural networks, and ensemble clustering. The approach substantially improves cell type-specific clusters when used with standard clustering algorithms like k-means and a state-of-the-art kernel-based clustering algorithm (SIMLR).

Topic

RNA-Seq;Machine learning;Transcriptomics;Cell biology

Detail

  • Operation: Principal component visualisation;Clustering;RNA-Seq analysis

  • Software interface: Command-line user interface

  • Language: R

  • License: -

  • Cost: Free

  • Version name: -

  • Credit: The Australian Research Council Discovery Early Career Researcher Award, National Health and Medical Research Council (NHMRC) Investigator Grant, the Australian Research Council Discovery Projects, the National Health and Medical Research Council (NHMRC)/Career Development Fellowship, the Australian Government Research Training Program Scholarship, the Judith and David Coffey Life Lab Gift scholarship.

  • Input: -

  • Output: -

  • Contact: Pengyi Yang pengyi.yang@sydney.edu.au

  • Collection: -

  • Maturity: -

Publications

  • Autoencoder-based cluster ensembles for single-cell RNA-seq data analysis.
  • Geddes TA, et al. Autoencoder-based cluster ensembles for single-cell RNA-seq data analysis. Autoencoder-based cluster ensembles for single-cell RNA-seq data analysis. 2019; 20:660. doi: 10.1186/s12859-019-3179-5
  • https://doi.org/10.1186/S12859-019-3179-5
  • PMID: 31870278
  • PMC: PMC6929272

Download and documentation

< Back to DB search