1
0
mirror of https://github.com/msberends/AMR.git synced 2024-12-26 21:26:12 +01:00
AMR/inst/tinytest/test-resistance_predict.R

96 lines
5.0 KiB
R

# ==================================================================== #
# TITLE #
# Antimicrobial Resistance (AMR) Data Analysis for R #
# #
# SOURCE #
# https://github.com/msberends/AMR #
# #
# LICENCE #
# (c) 2018-2022 Berends MS, Luz CF et al. #
# Developed at the University of Groningen, the Netherlands, in #
# collaboration with non-profit organisations Certe Medical #
# Diagnostics & Advice, and University Medical Center Groningen. #
# #
# This R package is free software; you can freely use and distribute #
# it for both personal and commercial purposes under the terms of the #
# GNU General Public License version 2.0 (GNU GPL-2), as published by #
# the Free Software Foundation. #
# We created this package for both routine data analysis and academic #
# research and it was publicly released in the hope that it will be #
# useful, but it comes WITHOUT ANY WARRANTY OR LIABILITY. #
# #
# Visit our website for the full manual and a complete tutorial about #
# how to conduct AMR data analysis: https://msberends.github.io/AMR/ #
# ==================================================================== #
if (AMR:::pkg_is_available("dplyr", min_version = "1.0.0")) {
expect_stdout(AMX_R <- example_isolates %>%
filter(mo == "B_ESCHR_COLI") %>%
rsi_predict(col_ab = "AMX",
col_date = "date",
model = "binomial",
minimum = 10,
info = TRUE) %>%
pull("value"))
# AMX resistance will increase according to data set `example_isolates`
expect_true(AMX_R[3] < AMX_R[20])
}
expect_stdout(x <- suppressMessages(resistance_predict(example_isolates,
col_ab = "AMX",
year_min = 2010,
model = "binomial",
info = TRUE)))
pdf(NULL) # prevent Rplots.pdf being created
expect_silent(plot(x))
if (AMR:::pkg_is_available("ggplot2")) {
expect_silent(ggplot_rsi_predict(x))
expect_silent(autoplot(x))
expect_error(ggplot_rsi_predict(example_isolates))
}
expect_stdout(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "binomial",
col_ab = "AMX",
col_date = "date",
info = TRUE))
expect_stdout(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "loglin",
col_ab = "AMX",
col_date = "date",
info = TRUE))
expect_stdout(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "lin",
col_ab = "AMX",
col_date = "date",
info = TRUE))
expect_error(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "INVALID MODEL",
col_ab = "AMX",
col_date = "date",
info = TRUE))
expect_error(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "binomial",
col_ab = "NOT EXISTING COLUMN",
col_date = "date",
info = TRUE))
expect_error(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "binomial",
col_ab = "AMX",
col_date = "NOT EXISTING COLUMN",
info = TRUE))
expect_error(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
col_ab = "AMX",
col_date = "NOT EXISTING COLUMN",
info = TRUE))
expect_error(rsi_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
col_ab = "AMX",
col_date = "date",
info = TRUE))
# almost all E. coli are MEM S in the Netherlands :)
expect_error(resistance_predict(x = subset(example_isolates, mo == "B_ESCHR_COLI"),
model = "binomial",
col_ab = "MEM",
col_date = "date",
info = TRUE))