tests/testthat/test_statistical_tests.R

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

test_that("test_houde works", {
    curves <- readRDS("noisy_curves.RDS")[[1]][[1]]

    expect_equal(test_houde(curves),
                 structure(list(Test = "Houde",
                                State_1 = structure(1L, .Label = c("10", "20"),
                                                    class = "factor"),
                                State_2 = structure(2L, .Label = c("10", "20"),
                                                    class = "factor"),
                                Test_statistic = NA,
                                P_value = NA,
                                Significant_difference = TRUE,
                                Time = NA,
                                Transformation = NA,
                                AIC = NA,
                                logLik = NA),
                           row.names = c(NA, -1L),
                           class = c("data.table", "data.frame")))
})


test_that("test_hdx_analyzer works", {
    curves <- readRDS("noisy_curves.RDS")[[1]][[1]]

    expect_equal(test_hdx_analyzer(curves),
                 structure(list(Test = c("Deuteros lm identity continuous",
                                         "Deuteros lm identity categorical",
                                         "Deuteros lm log continuous"),
                                State_1 = structure(c(1L, 1L, 1L),
                                                    class = "factor",
                                                    .Label = c("10", "20")),
                                State_2 = structure(c(2L, 2L, 2L),
                                                    class = "factor",
                                                    .Label = c("10", "20")),
                                Test_statistic = c(1.90596540277125,
                                                   30.4891372765257,
                                                   5.00315365023482),
                                P_value = c(0.151957845006165,
                                            3.63467693806125e-26,
                                            0.00777797073256792),
                                Significant_difference = c(FALSE, TRUE, TRUE),
                                Time = c("continuous",
                                         "categorical",
                                         "continuous"),
                                Transformation = c("identity",
                                                   "identity",
                                                   "log"),
                                AIC = c(168.992007835247,
                                        -404.05519750975,
                                        8.47535423232984),
                                logLik = c(-79.4960039176237,
                                           217.027598754875,
                                           0.762322883835081)),
                           row.names = c(NA, -3L),
                           class = c("data.table", "data.frame")))
})


test_that("test_memhdx_model works", {
    set.seed(10)
    curves <- readRDS("noisy_curves.RDS")[[1]][[1]]

    expect_equal(test_memhdx_model(curves),
                 structure(list(Test = c("MEMHDX lmm identity continuous",
                                         "MEMHDX lmm identity categorical",
                                         "MEMHDX lmm log continuous"),
                                State_1 = structure(c(1L, 1L, 1L),
                                                    class = "factor",
                                                    .Label = c("10", "20")),
                                State_2 = structure(c(2L, 2L, 2L),
                                                    class = "factor",
                                                    .Label = c("10", "20")),
                                Test_statistic = c(3.86021412427414,
                                                   148.417835464572,
                                                   9.94982009055384),
                                P_value = c(0.145132659439139,
                                            8.72405171449471e-29,
                                            0.00690914042131481),
                                Significant_difference = c(FALSE, TRUE, TRUE),
                                Time = c("continuous",
                                         "categorical",
                                         "continuous"),
                                Transformation = c("identity",
                                                   "identity",
                                                   "log"),
                                AIC = c(170.992007835247,
                                        -404.387996166469,
                                        10.4753542323301),
                                logLik = c(-79.4960039176237,
                                           218.193998083235,
                                           0.762322883834969)),
                           row.names = c(NA, -3L),
                           class = c("data.table", "data.frame")))
})


test_that("test_semiparametric works", {
    set.seed(10)
    curves <- readRDS("noisy_curves.RDS")[[1]][[1]]
    expect_equal(test_semiparametric(curves),
                 structure(list(Test = "Semiparametric model",
                                State_1 = structure(1L,
                                                    .Label = c("10", "20"),
                                                    class = "factor"),
                                State_2 = structure(2L,
                                                    .Label = c("10", "20"),
                                                    class = "factor"),
                                Test_statistic = 18.8790884443026,
                                P_value = 7.9516642216805e-05,
                                Significant_difference = TRUE,
                                Time = NA,
                                Transformation = "identity",
                                AIC = -367.11483443455,
                                logLik = 195.557417217275),
                           row.names = c(NA, -1L),
                           class = c("data.table", "data.frame")))
})

test_that("test_semiparametric works in the case or one or two time points", {
    set.seed(10)
    curves <- readRDS("noisy_curves.RDS")[[1]][[1]]
    curves <- curves[curves[["Exposure"]] %in% c(5, 60)]
    expect_equal(test_semiparametric(curves),
                 structure(list(Test = "Semiparametric model",
                                State_1 = structure(1L,
                                                    .Label = c("10", "20"),
                                                    class = "factor"),
                                State_2 = structure(2L,
                                                    .Label = c("10", "20"),
                                                    class = "factor"),
                                Test_statistic = 46.9115145699878,
                                P_value = 6.50570014465104e-11,
                                Significant_difference = TRUE,
                                Time = NA, Transformation = "identity",
                                AIC = -121.276210144012,
                                logLik = 67.6381050720061),
                           row.names = c(NA, -1L),
                           class = c("data.table", "data.frame")))
})



test_that("truncated_lines works", {
    set.seed(10)
    x <- rnorm(30)
    knots <- seq(min(x), max(x), 1)
    expect_equal(truncated_lines(x, knots),
                 structure(c(2.20403300911136, 2.00103429610047,
                             0.813956288247019, 1.58611912238581,
                             2.47983196473704, 2.5750811388697,
                             0.977210662740044, 1.82161082069867,
                             0.55861415646644, 1.92880844404554,
                             3.28706634125666, 2.94106834619687,
                             1.94705328215081, 3.17273154158292,
                             2.92667696655335, 2.27463410466535,
                             1.23034298201715, 1.99013645350229,
                             3.11080810026361, 2.66826536300614,
                             1.58897620144932, 0, 1.51042090029441,
                             0.0662256462593565, 0.920088816638627,
                             1.81162528301483, 1.49773140778161,
                             1.31312801145184, 2.08352583194471,
                             1.93150630806707, 1.20403300911136,
                             1.00103429610047, 0,
                             0.586119122385812, 1.47983196473704,
                             1.5750811388697, 0, 0.821610820698668,
                             0, 0.928808444045539, 2.28706634125666,
                             1.94106834619687, 0.947053282150812,
                             2.17273154158292, 1.92667696655335,
                             1.27463410466535, 0.230342982017153,
                             0.990136453502291, 2.11080810026361,
                             1.66826536300614, 0.588976201449324,
                             0, 0.510420900294414, 0, 0,
                             0.811625283014828, 0.497731407781613,
                             0.313128011451841, 1.08352583194471,
                             0.931506308067068, 0.204033009111357,
                             0.00103429610046663, 0, 0,
                             0.479831964737038, 0.575081138869698,
                             0, 0, 0, 0, 1.28706634125666,
                             0.941068346196868, 0, 1.17273154158292,
                             0.926676966553354, 0.274634104665352,
                             0, 0, 1.11080810026361, 0.668265363006141,
                             0, 0, 0, 0, 0, 0, 0, 0, 0.0835258319447139,
                             0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                             0.287066341256661, 0, 0, 0.17273154158292,
                             0, 0, 0, 0, 0.11080810026361, 0, 0,
                             0, 0, 0, 0, 0, 0, 0, 0, 0
                 ), .Dim = c(30L, 4L)))
})