Construct a list of input data — constructDataPop • scDesignPop

Function extracts an expression matrix, cell covariates, and filters the SNPs in the eQTL genotype dataframe.

Usage

constructDataPop(
  sce,
  eqtlgeno_df,
  new_eqtlgeno_df = NULL,
  new_covariate = NULL,
  overlap_features = NULL,
  sampid_vec = NULL,
  copula_variable = celltype_colname,
  n_quantiles = 10,
  slot_name = "counts",
  snp_mode = c("single", "multi"),
  time_colname = NULL,
  celltype_colname = "cell_type",
  feature_colname = "gene_id",
  snp_colname = "snp_id",
  loc_colname = "POS",
  chrom_colname = "CHR",
  indiv_colname = "indiv",
  prune_thres = 0.9,
  ...
)

Arguments

sce: a SingleCellExperiment object.
eqtlgeno_df: a dataframe with eQTL annotations and SNP genotypes for each gene.
new_eqtlgeno_df: a dataframe with eQTL annotations and SNP genotypes for each gene in new individuals. The default is NULL.
new_covariate: a cell covariate dataframe for which to simulate. The default is NULL.
overlap_features: an optional string vector to filter for features (ie. genes). The default is NULL.
sampid_vec: an optional string vector to filter for sample ids. The default is NULL.
copula_variable: a string scalar specifying the column name in cell covariate of sce object for which to fit the copula by. If NULL, then copula will be fitted for all cells. if it's the same as celltype_colname, then copula will be fitted per cell type, while if it's the same as time_colname, copula will be fitted per time quantile.
n_quantiles: an integer value used for how many time quantiles of continuous cell states will be assigned for copula fitting. The default is 10.
slot_name: a string scalar specifying the slot to use in input sce. The default is "counts".
snp_mode: a string scalar specifying the type of SNP mode used in the marginal models. Options are either "single" for single-SNP mode, or "multi" for multi-SNP mode. If multi-SNP mode is used, each gene will have one or more SNPs in the model upon SNP pruning. The default is "single".
time_colname: a string scalar specifying the time/pseudotime variable in the cell covariate of sce object.
celltype_colname: a string scalar specifying the cell type variable in eqtlgeno_df and cell covariate of sce object. The default is "cell_type".
feature_colname: a string scalar specifying the feature variable (ie. genes) in eqtlgeno_df. The default is "gene_id".
snp_colname: a string scalar for the SNP variable in eqtlgeno_df. The default is "snp_id".
loc_colname: a string scalar for the last column of eQTL annotation in eqtlgeno_df. The default is "POS".
chrom_colname: a string scalar of the chromosome variable in eqtlgeno_df. The default is "CHR".
indiv_colname: a string scalar of the sample ID variable in cell covariate of sce. The default is "indiv".
prune_thres: a numerical value between 0 and 1 used to threshold the pairwise correlations of eQTLs' genotypes for each feature. The default value is 0.9.
...: additional arguments passed to internal functions.

Value

outputs a list with following elements:

count_mat: a cell-by-gene matrix of response values.
covariate: a cell-by-covariate data frame used for fit marginal.
new_covariate: an optional cell-by-covariate data frame used in simulation.
important_features: a string vector of gene ids.
noeqtl_features: a string vector of gene ids that do not have eQTLs.
eqtl_geno_list: a list of eQTL genotype dataframes for each gene (for fit marginal).
new_eqtl_geno_list: a optional list of eQTL genotype dataframes for each gene (for new individual simulation).
filtered_gene: string vector of features QC filtered.

Examples

NULL
#> NULL