calculate_hdx_power: Calculate power of statistical tests for HDX experiments

In hadexversum/powerHDX: Efficient Simulation of HDX-MS Data and Tools for the Statistical Analysis

Calculate power of statistical tests for HDX experiments

Description

This function estimates power of statistical tests for HDX experiments.

Usage

calculate_hdx_power(
  deuteration_curves,
  tests,
  significance_level = 0.05,
  summarized = TRUE
)

Arguments

deuteration_curves	list returned by the get_noisy_deuteration_curves
tests	lists of tests to perform. Each test function should take two parameters - data (data_table containing replicated curves) and significance_level, and have particular output - data frame of variables: Test (name of a test which should be displayed in the final result), State_1, State_2 (biological states of interest), Test_statistic, P_value, Significant_difference (the same as p_value <= significance_level), Time (character, "continuous" or "categorical"), Transformation (character, transformation that is used for exposure), AIC, logLik. For example see test_houde.
significance_level	significance level that will be used for testing. See tests
summarized	logical. Indicates whether the power should be calculated. Default TRUE.

Value

list of data.tables with test result, optionally summarized with power.

See Also

test_houde, test_semiparametric, test_hdx_analyzer, test_memhdx_model

Examples

theo_spectra_pf_100 <- simulate_theoretical_spectra(sequence = "LVRKDLQN",
                                                    charge = c(3, 5),
                                                    protection_factor = 100,
                                                    times = c(0.167, 5),
                                                    pH = 7.5,
                                                    temperature = 15,
                                                    n_molecules = 500,
                                                    time_step_const = 1,
                                                    use_markov = TRUE)

theo_spectra_pf_200 <- simulate_theoretical_spectra(sequence = "LVRKDLQN",
                                                    charge = c(3, 5),
                                                    protection_factor = 200,
                                                    times = c(0.167, 5),
                                                    pH = 7.5,
                                                    temperature = 15,
                                                    n_molecules = 500,
                                                    time_step_const = 1,
                                                    use_markov = TRUE)

theo_spectra_two_states <- rbind(theo_spectra_pf_100, theo_spectra_pf_200)

deut_curves_p_states <- get_noisy_deuteration_curves(theo_spectra_two_states,
                                                     n_replicates = 4,
                                                     n_experiments = 2,
                                                     compare_pairs = TRUE,
                                                     reference = "all")
calculate_hdx_power(deut_curves_p_states,
                    tests = list(test_houde),
                    summarized = TRUE)

hadexversum/powerHDX documentation built on Aug. 31, 2022, 7:47 p.m.