Convert all character to factor

convert_character_to_factor.R

iris_e <- iris
str(iris_e)
# from factor to character
iris_e <- iris_e %>% dplyr::mutate(tidyselect::across(tidyselect::where(is.factor), as.character))
str(iris_e)
# from character to factor
iris_e <- iris_e %>% dplyr::mutate(tidyselect::across(tidyselect::where(is.character), as.factor))
str(iris_e)

# remove all-NA columns
df2 %>% dplyr::select(tidyselect::where(function(x) any(!is.na(x))))


## Standardize (selected) numeric variables
list_var_standardize <-
  c(
    "Abeta_1_42"
  #, "ApoE"
  , "Alpha_synuclein"
  , "p_Tau"
  , "t_Tau"
  , "p_Tau_log2"
  , "t_Tau_log2"
  , Var[["list"]][["var_vol"]]
  , Var[["list"]][["var_cog"]]
  #, "ICV"
  )

# means subtracted to center
dat_sub_ic %>%
  dplyr::select(
    ICV
  , age_visit
  , EDUCYRS
  ) %>%
  dplyr::summarize(
    ICV       = (ICV / 1e6) %>% mean(na.rm = TRUE)
  , age_visit = age_visit %>% mean(na.rm = TRUE)
  , EDUCYRS   = EDUCYRS   %>% mean(na.rm = TRUE)
  )

dat_sub_ic_z <-
  dat_sub_ic %>%
  dplyr::mutate(
    # z-score vol and cog vars
    across(
      .cols = all_of(list_var_standardize)
    , .fns  = ~ scale(x = ., center = TRUE, scale = TRUE)[,1]
    )
    # center demographic variables
  #, ICV       = scale(ICV / 1e6, center = TRUE, scale = FALSE) %>% as.numeric()
  #, age_visit = scale(age_visit, center = TRUE, scale = FALSE) %>% as.numeric()
  , ICV       = scale(ICV / 1e6, center = TRUE, scale = TRUE) %>% as.numeric()
  , age_visit = scale(age_visit, center = TRUE, scale = TRUE) %>% as.numeric()
    # 12/18/2021 6:19PM comment out to interpret plot
  #, EDUCYRS   = scale(EDUCYRS  , center = TRUE, scale = FALSE) %>% as.numeric()
  , EDUCYRS   = scale(EDUCYRS  , center = TRUE, scale = TRUE) %>% as.numeric()
  , GENDER    = GENDER - 1  # females are baseline
  , NP3TOT    = scale(NP3TOT, center = TRUE, scale = TRUE) %>% as.numeric()
  , LEDD_total = scale(LEDD_total, center = FALSE, scale = TRUE) %>% as.numeric()
  )