Eval_multi: Evaluate estimation methods
In xuranw/MuSiC: Multi-subject single cell deconvolution

Eval_multi

R Documentation

Evaluate estimation methods

Calculate Pearson correlation, RMSE and mAD between real and estimated cell type proportions

Eval_multi(
  prop.real,
  prop.est,
  method.name = NULL,
  by.subject = FALSE,
  select.ct = NULL,
  ...
)

`prop.real`	a matrix of real cell type proportions
`prop.est`	a matrix or a list of matrices of estimated cell type proportions
`method.name`	vector of the names of estmation methods. Default is NULL and the names will be generated automatically as 'Est1', 'Est2', ...
`by.subject`	logical, default is FALSE. If TRUE, Pearson correlation is estimated subject by subject.
`select.ct`	a vector of cell types selected for evaluation. Default is NULL. If select.ct is NULL, we evaluate real and estimated cell type proportion with all common cell types. Otherwise use only selected cell types.