conos

Conos R6 class

Create and preprocess a Seurat object

Find threshold of cluster detectability

Find threshold of cluster detectability in trees of clusters

Rescale the weights in an edge matrix to match a given perplexity.

Check that the count data contain only integer counts

Collapse count matrices by cell type, given min/max number of cells

Convert Conos object to Pagoda2 object

Set edge matrix edgeMat with certain values on sample

Estimate entropy of edge weights per cell according to the specified factor. Can be used to visualize alignment quality according to this factor.

Increase resolution for a specific set of clusters

Compare two cell types across the entire panel

Compare two cell types across the entire panel

Access cell names from sample

Evaluate consistency of cluster relationships Using the clustering we are generating per-sample dendrograms and we are examining their similarity between different samples More information about similarity measures <https://www.rdocumentation.org/packages/dendextend/versions/1.8.0/topics/cor_cophenetic> <https://www.rdocumentation.org/packages/dendextend/versions/1.8.0/topics/cor_bakers_gamma>

Access clustering from sample

Extract specified clustering from list of conos clusterings

Access count matrix from sample

Access embedding from sample

Access gene expression from sample

Access genes from sample

Deprecated; Get markers for global clusters

Establish rough neighbor matching between samples given their projections in a common space

Get top overdispersed genes across samples

Access overdispersed genes from sample

Access PCA from sample

Do differential expression for each cell type in a conos object between the specified subsets of apps

Evaluate how many clusters are global

Access raw count matrix from sample

Retrieve sample names per cell

Performs a greedy top-down selective cut to optmize modularity

Constructrs a two-step clustering, first running multilevel.communities, and then walktrap.communities within each These are combined into an overall hierarchy

Constructs a two-step clustering, first running multilevel.communities, and then walktrap.communities within each These are combined into an overall hierarchy

Get a vector with the levels of a factor named with their own name. Useful for lapply loops over factor levels

Get a vector of the names of an object named by the names themselves. This is useful with lapply when passing names of objects as it ensures that the output list is also named

Utility function to generate a pagoda2 app from a conos object

Plots barplots per sample of composition of each pagoda2 application based on selected clustering

Generate boxplot per cluster of the proportion of cells in each celltype

Plot fraction of variance explained by the successive reduced space components (PCA, CPCA)

Plot a heatmap of differential genes

Plot panel of specified embeddings

Plot panel of specified embeddings, extracting them from pagoda2 objects

Project a distance matrix into a lower-dimensional space.

Estimate labeling distribution for each vertex, based on provided labels using Random Walk

Perform CCA (using PMA package or otherwise) on two samples

Perform cpca on two samples

Use space of combined sample-specific PCAs as a space

Get raw matrices with common genes

Objects exported from other packages

Save Conos object on disk to read it from ScanPy

Save differential expression as table in *csv format

Save differential expression results as JSON

Scan joint graph modularity for a range of k (or k.self) values Builds graph with different values of k (or k.self if scan.k.self=TRUE), evaluating modularity of the resulting multilevel clustering NOTE: will run evaluations in parallel using con$n.cores (temporarily setting con$n.cores to 1 in the process)

Calculate the default number of batches for a given number of vertices and edges. The formula used is the one used by the 'largeVis' reference implementation. This is substantially less than the recommendation E * 10000 in the original paper.

Small pre-processed data from Pagoda2, two samples, each dimension (1000, 100)

Determine number of detectable clusters given a reference walktrap and a bunch of permuted walktraps