process_dgTMatrix_lists: Count Matrix To Signature Matrix
In DustinSokolowski/scMappR: Single Cell Mapper

Description Usage Arguments Details Value Examples

View source: R/process_dgTMatrix_lists.R

This function takes a list of count matrices, processes them, calls cell-types, and generates signature matrices.

process_dgTMatrix_lists(
  dgTMatrix_list,
  name,
  species_name,
  naming_preference = -9,
  rda_path = "",
  panglao_set = FALSE,
  haveUMAP = FALSE,
  saveSCObject = FALSE,
  internal = FALSE,
  toSave = FALSE,
  path = NULL,
  use_sctransform = FALSE,
  test_ctname = "wilcox",
  genes_integrate = 2000,
  genes_include = FALSE
)

`dgTMatrix_list`	A list of matrices in the class of dgTMatrix object – sparce object – compatible with Seurat rownames should be of the same species for each.
`name`	The name of the outputted signature matrices, cell-type preferences, and seurat objects if you choose to save them.
`species_name`	Mouse or human symbols, -9 if internal as panglao objects have gene symbol and ensembl strapped together.
`naming_preference`	For cell-type naming, see if cell-types given the inputted tissues are more likely to be named within one of the categories of get_naming_preference_options().
`rda_path`	If saved, directory to where data from scMappR_data is downloaded.
`panglao_set`	If the inputted matrices are from Panglao (i.e. if they're internal).
`haveUMAP`	Save the UMAPs – only use if the package is downloaded with pip.
`saveSCObject`	Save the Seurat object as an RData object (T/F).
`internal`	Was this used as part of the internal processing of Panglao datasets (T/F).
`toSave`	Allow scMappR to write files in the current directory (T/F)
`path`	If toSave == TRUE, path to the directory where files will be saved.
`use_sctransform`	If you should use sctransform or the Normalize/VariableFeatures/ScaleData pipeline (T/F).
`test_ctname`	statistical test for calling CT markers – must be in Seurat
`genes_integrate`	The number of genes to include in the integration anchors feature when combining datasets – passed into process_from_count
`genes_include`	TRUE or FALSE – include 2000 genes in signature matrix or all matrix.

This function is a one line wrapper to process count matrices into a signature matrix. It combines process_from_count, two methods of identifying cell-type identities (GSVA and Fisher's test). Then, it takes the output of cell-type markers and converts it into a signature matrix of p-value ranks and odds ratios. Along the way, it saves the Seurat object (if chosen with saveSCObject), cell-type identites from GSVA (its own obect), and the signature matrices. Cell-type marker outputs are also saved in the generes.RData list. Names of the generes objects and the signature matrices are kept.

List with the following elements:

`wilcoxon_rank_mat_t`	A dataframe containing the signature matrix of ranks (-log10(Padj) * sign(fold-change)).
`wilcoxon_rank_mat_or`	A dataframe containing the signature matrix of odds-ratios.
`generes`	All cell-type markers for each cell-type with p-value and fold changes.

data(sm)
toProcess <- list(example = sm)
tst1 <- process_dgTMatrix_lists(toProcess, name = "testPropcess", species_name = -9,
 naming_preference = "eye", rda_path = "")