p2app4conos: Utility function to generate a pagoda2 app from a conos...

In conos: Clustering on Network of Samples

Description

Utility function to generate a pagoda2 app from a conos object

Usage

p2app4conos(
  conos,
  cdl = NULL,
  metadata = NULL,
  filename = "conos_app.bin",
  save = TRUE,
  n.cores = 1,
  n.odgenes = 3000,
  nPcs = 100,
  k = 30,
  perplexity = 50,
  log.scale = TRUE,
  trim = 10,
  keep.genes = NULL,
  min.cells.per.gene = 0,
  min.transcripts.per.cell = 100,
  get.largevis = TRUE,
  get.tsne = TRUE,
  make.geneknn = TRUE,
  go.env = NULL,
  cell.subset = NULL,
  max.cells = Inf,
  additional.embeddings = NULL,
  test.pathway.overdispersion = FALSE,
  organism = NULL,
  return.details = FALSE
)

Arguments

conos	Conos object
cdl	list Optional list of raw matrices (so that gene merging doesn't have to be redone) (default=NULL)
metadata	list Optional list of (named) metadata factors (default=NULL)
filename	string Name of the *.bin file to seralize for the pagoda2 application if save=TRUE (default='conos_app.bin')
save	boolean Save serialized *bin file specified in filename (default=TRUE)
n.cores	integer Number of cores (default=1)
n.odgenes	numeric Number of top overdispersed genes to use (dfault=3e3). From pagoda2::basicP2proc().
nPcs	numeric Number of PCs to use (default=100). From pagoda2::basicP2proc().
k	numeric Default number of neighbors to use in kNN graph (default=30). From pagoda2::basicP2proc().
perplexity	numeric Perplexity to use in generating tSNE and largeVis embeddings (default=50). From pagoda2::basicP2proc().
log.scale	boolean Whether to use log scale normalization (default=TRUE). From pagoda2::basicP2proc().
trim	numeric Number of cells to trim in winsorization (default=10). From pagoda2::basicP2proc().
keep.genes	optional set of genes to keep from being filtered out (even at low counts) (default=NULL). From pagoda2::basicP2proc().
min.cells.per.gene	numeric Minimal number of cells required for gene to be kept (unless listed in keep.genes) (default=0). From pagoda2::basicP2proc().
min.transcripts.per.cell	numeric Minimumal number of molecules/reads for a cell to be admitted (default=100). From pagoda2::basicP2proc().
get.largevis	boolean Whether to caluclate largeVis embedding (default=TRUE). From pagoda2::basicP2proc().
get.tsne	boolean Whether to calculate tSNE embedding (default=TRUE). From pagoda2::basicP2proc().
make.geneknn	boolean Whether pre-calculate gene kNN (for gene search) (default=TRUE). From pagoda2::basicP2proc().
go.env	GO environment for the organism of interest (default=NULL)
cell.subset	string Cells to subset with the conos embedding conos$embedding. If NULL, uses all cells via rownames(conos$embedding) (default=NULL)
max.cells	numeric Limit to the cells that are included in the conos. If Inf, there is no limit (default=Inf)
additional.embeddings	list Additional embeddings to add to conos for the pagoda2 app (default=NULL)
test.pathway.overdispersion	boolean Find all IDs using GO category against either org.Hs.eg.db ('hs') or org.Mm.eg.db ('mm') (default=FALSE
organism	string Organism of interest, either 'hs' (Homo sapiens) or 'mm' (Mus musculus, i.e. mouse) (default=NULL). Only used if test.pathway.overdispersion is TRUE. If NULL and test.pathway.overdispersion=TRUE, then 'hs' is used.
return.details	boolean If TRUE, return list of p2 application, pagoda2 object, list of raw matrices, and cell names. If FALSE, simply return pagoda2 app object. (default=FALSE)

Value

pagoda2 app object

Examples

 
library(pagoda2)
panel.preprocessed <- lapply(conosPanel::panel, basicP2proc, n.cores=1, min.cells.per.gene=0, 
    n.odgenes=2e3, get.largevis=FALSE, make.geneknn=FALSE)
con <- Conos$new(panel.preprocessed, n.cores=1)
con$buildGraph(k=30, k.self=5, space='PCA', ncomps=30, n.odgenes=2000, matching.method='mNN', 
    metric='angular', score.component.variance=TRUE, verbose=TRUE)
con$findCommunities(method=leiden.community, resolution=1)
con$embedGraph(alpha=0.001, sgd_batched=1e8) 
p2app4conos(con)

conos documentation built on Feb. 23, 2021, 9:06 a.m.