jrSiCKLSNMF

This package contains code to run joint graph-regularized single-cell Kullback-Leibler Non-negative Matrix Factorization

To install:

install.packages("devtools")
devtools::install_github("ellisdoro/jrSiCKLSNMF")

Copy Link

Version

Install

install.packages('jrSiCKLSNMF')

Monthly Downloads

203

Version

1.2.3

License

GPL-3

Maintainer

Dorothy Ellis

Last Published

July 11th, 2025

Functions in jrSiCKLSNMF (1.2.3)

SickleJr-class

The SickleJr class

SetWandHfromWHinitials

Set \(\mathbf{W}\) matrices and \(\mathbf{H}\) matrix from pre-calculated values

PlotSickleJrUMAP

Generate UMAP plots for an object of class SickleJr

RunjrSiCKLSNMF

Run jrSiCKLSNMF on an object of class SickleJr

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.

jrSiCKLSNMF

Run jrSiCKLSNMF outside of a SickleJr object

MinibatchDiagnosticPlot

Plot a diagnostic plot for the mini-batch algorithm

BuildSNNGraphLaplacians

Build SNN graphs and generate their graph Laplacians

ClusterSickleJr

Cluster the \(\mathbf{H}\) matrix

BuildKNNGraphLaplacians

Build KNN graphs and generate their graph Laplacians

DetermineClusters

Perform clustering diagnostics

GenerateWmatricesandHmatrix

Initialize the \(\mathbf{W}\) matrices in each modality and the shared \(\mathbf{H}\) matrix

CreateSickleJr

Create an object of class SickleJr

DetermineDFromIRLBA

Create elbow plots of the singular values derived from IRLBA to determine D for large datasets

AddSickleJrMetadata

Add metadata to an object of class SickleJr

CalculateUMAPSickleJr

Calculate the UMAP for an object of class SickleJr

SetLambdasandRowReg

Set lambda values and type of row regularization for an object of class SickleJr

SimData

A simulated dataset for use with jrSiCKLSNMF

PlotLossvsLatentFactors

Create plots to help determine the number of latent factors

NormalizeCountMatrices

Normalize the count matrices and set whether to use the Poisson KL divergence or the Frobenius norm