In InfOmics/ISDBSCAN-R: Clustering Large Single-Cell Datasets

Installation

To install the package, please use the following.

devtools::install_github("drisso/mbkmeans")

Introduction

This vignette provides an introductory example on how to work with the ISDBSCAN package, which contains an implementation of the density based algorithm (ISDBSCAN) for large single-cell sequencing data.

The main function to be used by the users is ISDBSCAN. This is implemented as an S4 generic and methods are implemented for matrix, Matrix, HDF5Matrix.

Our main contribution here is to provide an interface to the DelayedArray and HDF5Array packages, allowing the user to run the algorithm on data that do not fit entirely in memory.

The motivation for this work is the clustering of large single-cell RNA-sequencing (scRNA-seq) datasets, and hence the main focus is on Bioconductor's SingleCellExperimentand SummarizedExperiment data container. For this reason, ISDBSCAN assumes a data representation typical of genomic data, in which genes (variables) are in the rows and cells (observations) are in the column. This is contrary to most other statistical applications.

This algorithm can be run with a preprocesing, called stratification stratif=TRUE. This procedure will induce an order in the dataset based on the influence of the points. When the analized datasets are very large, and don't fit into the RAM memory, it would be better to avoid this tipe of preprocesing.