Skip to contents

apply_fit_to_se

Source: R/fit_utils.R
apply_fit_to_se.Rd

Apply a user-supplied fit function per (drug x cell line x normalization_type) triplet and persist results into the Metrics assay.

Usage

apply_fit_to_se(
  se,
  fit_fn,
  normalization_types = c("GR", "RV"),
  averaged_assay = "Averaged",
  metrics_assay = "Metrics",
  merge = "merge",
  on_error = "warn",
  fit_source = "custom"
)

Arguments

se

SummarizedExperiment with an Averaged assay

fit_fn

function taking a data.table and returning a named list

normalization_types

character vector of types to iterate

averaged_assay

name of the input assay

metrics_assay

name of the output assay

merge

"merge" (idempotent upsert) or "replace"

on_error

"warn" (skip + warning) or "stop"

fit_source

string recorded in the fit_source column

Value

updated SummarizedExperiment

Details

This is a convenience wrapper around apply_fit for the single-agent use case with the standard Metrics assay.

Examples

mae <- gDRutils::get_synthetic_data("finalMAE_small.qs2")
se <- mae[["single-agent"]]
simple_fn <- function(dt) list(x_mean = mean(dt$x, na.rm = TRUE))
se_out <- apply_fit_to_se(se, simple_fn, fit_source = "demo")
#> Warning: fit_fn output missing columns: x_mean_sd, x_AOC_sd, x_AOC_range_sd, xc50_sd, x_max_sd, ec50_sd, x_inf_sd, x_0_sd, h_sd, r2_sd, x_sd_avg_sd