jrSiCKLSNMF (version 1.2.3)

SimSickleJrSmall: A small SickleJr object containing a subset of data from the SimData data object. Contains the completed analysis from the `Getting Started` vignette for a small subset of 10 cells with 150 genes and 700 peaks. The clusters derived from this dataset are not accurate; this dataset is intended for use with code examples.

Description

A small SickleJr object containing a subset of data from the SimData data object. Contains the completed analysis from the `Getting Started` vignette for a small subset of 10 cells with 150 genes and 700 peaks. The clusters derived from this dataset are not accurate; this dataset is intended for use with code examples.

Usage

data(SimSickleJrSmall)

Format

A SickleJr object containing a completed analysis using jrSiCKLSNMF

count.matrices

Contains a list of 2 sparse matrices, each containing a different simulated omics modality measured on the same set of single cells

normalized.count.matrices

The normalized versions of the count matrices contained in slot count.matrices

graph.laplacian.list

A list of sparse matrices containing the graph Laplacians corresponding to the KNN feature-feature similarity graphs constructed for each omics modality

rowRegularization

A string indicating the row regularization: here it is set to "None"

diffFunc

A string specifying the function to measure the discrepancy between the normalized data and the fitted matrices: here, it is set to "klp" for the Poisson Kullback-Leibler divergence

lambdaWlist

A list holding the graph regularization parameters: here, they are 10 and 50

lambdaH

A numeric indicating the value for the sparsity parameter. Here it is equals 500

Wlist

A list holding the fitted \(\mathbf{W}^v\) matrices

H

A matrix holding \(\mathbf{H}\)

WHinitials

A list of initial values for \(\mathbf{W}^v\) and \(\mathbf{H}\)

lossCalcSubsample

A vector containing a subset on which to calculate the loss

latent.factor.elbow.values

A data frame holding the loss and the number of latent factor that is used for diagnostic plots

minibatch

A Boolean indicating whether or not to use the mini-batch algorithm: FALSE here

clusterdiagnostics

Diagnostic plots and results

clusters

A list holding the "kmeans" clustering results

metadata

A list holding metadata; here this is just cell type information

loss

A list holding a vector called "Loss"

umap

A list holding various UMAP approximations

plots

A list holding ggplots corresponding to different diagnostics and visualizations