Linear mixed-effects model for brain imaging data.

Linear regression for brain imaging data.

bstr_lmer(
  group_var,
  main_effect = "",
  covariates = "",
  bstr_data,
  mult_comp = "fdr"
)

Arguments

group_var

Categorical variable name. This should be present in the demographics csv file associated with bstr_data.

main_effect

Character string containing an independent variable whose effect you want to measure. It could be disease status, age, gender etc. This should strictly be a single variable. This can be either a categorical or a continuous variable.

covariates

Character string containing a set of other predictors (variables) in the model. If more than one covariates are included, they should be separated by a + operator similar to an R formula.

bstr_data

Object of type BstrData()

mult_comp

method for multiple comparisons correction. The default method is "fdr". See bstr_p_adjust() for valid values.

Details

This function accepts a main_effect and a set of covariates (using the R formula notation) and performs a linear regression including main_effect + covariates.

Slightly different from the standard R lmer function, bstr_lmer currently does not directly accept an R formula. This could be accomodated in the future versions. Also currently, this function returns the p-values and the t-statistics for the main_effect only. Returning the statistics for all variables could be accomodated in the future versions.