logistic_prediction_cell: logistic_prediciton_cell

In hemberg-lab/sctransfermap: sctransfermap for predicting cell type in single-cell data (RNA-seq and ATAC-seq)

Description

predict cell type with logistic regression model

Usage

logistic_prediction_cell(iscell, meta_model_list, query,
  gn_overlap_thr = 0.5, logistic_thr = 0.5, tissue_opt = FALSE,
  celltype_specific = FALSE)

Arguments

iscell	attribute set
meta_model_list	result from create_meta_index
query	signature object of query
gn_overlap_thr	minimum gene overlap ratio of attribute in attribute set between query data
logistic_thr	threshold for each cell whether predicted score from logistic regression model is confident
tissue_opt	TRUE if user wants to find tissue-origin
celltype_specific	TRUE if user wants to calculate the weight fo each tissue with only matched cell type. Otherwise, weight for each tissue will be calculated with whole celltypes (default: FALSE; tissue_opt should be TRUE to use this parameter)

Value

return 2 confusion matrix

• confusion: a confusion matrix of prediction result

• all_confusion: a confusion matrix of prediction result in each cell

Examples

meta_model <- create_meta_index_av(att_set, sig_set)
res_list <- logistic_prediction_cell(att_set, meta_model, query_signature)
res_list$confusion or res_list$all_confusion

hemberg-lab/sctransfermap documentation built on Nov. 11, 2019, 6:23 a.m.