multi_fitteds: Get the fitted values from multiple linear models.
In mirvie/mirmodels: Models Built by the Informatics Team at Mirvie

Description Usage Arguments Value Examples

Get the fitted values from multiple linear models.

1	multi_fitteds(lms, lms_data = NULL, new_data = NULL)

`lms`	A list of fitted linear models (the results of calls to `stats::lm()` or `MASS::rlm()`). Most likely, the output of a call to `multi_lm()`.
`lms_data`	The data frame that was passed as the `df` argument to `multi_lm()`, if `multi_lm()` was used to create `lms`.
`new_data`	Rather than calculating the residuals on the data where the model was fit, you can pass a new dataset and calculate the residuals there.

A data frame of the fitted values The variables in this data frame will have their original names. If lms_data is given, other columns (ones for which residuals were not calculated) of that data frame will be returned as is.

if (rlang::is_installed("mirmisc")) {
  gars_data <- get_combined_cohort_data(c("ga", "rs"), log2 = TRUE) %>%
    dplyr::mutate(
      cohort = factor(
        dplyr::if_else(startsWith(mirvie_id, "RS"), "RS", "GA")
      ),
      meta_major_race = forcats::fct_drop(meta_major_race)
    ) %>%
    dplyr::filter(!is.na(meta_major_race), !is.na(meta_collectionga))
  xs <- c("cohort", "meta_major_race", "meta_collectionga")
  ys <- gars_data %>%
    dplyr::select(dplyr::any_of(mirmisc::get_gene_names())) %>%
    purrr::map_dbl(sum) %>%
    sort() %>%
    tail(100) %>%
    names() # ys are the highest expressed 100 genes
  lms <- multi_lm(gars_data, xs, ys)
  fitteds <- multi_fitteds(lms, gars_data)
}