Model Fitting

This function is used to fit the generalized linear mixed-effect model through nonparametric maximum likelihood estimation for a given gene.

modelFit(
  dat_phase,
  n_condition = "one",
  resampled_data = FALSE,
  varList = NULL
)

Arguments

dat_phase:

Psudo-phased bulk RNA-seq data of a given gene
n_condition:

a character string indicates whether to perform one condition analysis or two conditions analysis. Possible values are "one" or "two". Default is "one"
resampled_data:

a logical value indicates whether the data analyzed is obtained from resampling or not. Used only for two conditions analysis. Default is FALSE
varList:

a character string specifies fomula of covariates that users want to adjusted in the model, e.g., "`var1`+`var2`". Default is NULL

Value

  • One condition analysis: likelihood ratio test (LRT) statistic;

  • Two conditions analysis:
    - original data: estimated ASE effect in the pooled sample & likelihood ratio test (LRT) statistic;
    - resampled data: likelihood ratio test (LRT) statistic;