tests/testthat/test_normalization.R

In glycanr: Tools for Analysing N-Glycan Data

library(dplyr)
context("Normalization functions")

#library(dplyr)
#library(tidyr)
#library(glycanr)
#data(mpiu)
#
#source("../../..//normalizationFunctions_23062015.r")
#
#mw <- mpiu %>%
#    spread(glycan, value)
#
##
## tanorm
##
#tmp1 <- totalAreaNorm(mw) %>%
#    select(-normalization)
#
#tmp2 <- tanorm(mpiu) %>%
#    spread(glycan, value)
#tmp2 <- data.frame(tmp2)
#
#print(all.equal(tmp1, tmp2))
#if(all.equal(tmp1, tmp2)){
#    mpiunorm <- tmp2 %>%
#        mutate(norm="tanorm")
#}
#
#
#
##
## refPeakNorm
##
#tmp1 <- refPeakNorm(mw) %>%
#    select(-normalization)
#
#tmp2 <- refpeaknorm(mpiu) %>%
#    spread(glycan, value)
#tmp2 <- data.frame(tmp2)
#
#print(all.equal(tmp1, tmp2))
#if(all.equal(tmp1, tmp2)){
#    mpiunorm <- bind_rows(mpiunorm,
#                          tmp2 %>% mutate(norm="refpeaknorm"))
#}
#
#
##
## mediannorm
##
#tmp1 <- medianNorm(mw) %>%
#    select(-normalization)
#
#tmp2 <- mediannorm(mpiu) %>%
#    spread(glycan, value)
#tmp2 <- data.frame(tmp2)
#
#print(all.equal(tmp1, tmp2))
#if(all.equal(tmp1, tmp2)){
#    mpiunorm <- bind_rows(mpiunorm,
#                          tmp2 %>% mutate(norm="mediannorm"))
#}
#
##
## medianQuotientNorm
##
#tmp1 <- medianQuotientNorm(mw) %>%
#    select(-normalization)
#
#tmp2 <- medianquotientnorm(mpiu) %>%
#    spread(glycan, value)
#tmp2 <- data.frame(tmp2)
#
#print(all.equal(tmp1, tmp2))
#if(all.equal(tmp1, tmp2)){
#    mpiunorm <- bind_rows(mpiunorm,
#                          tmp2 %>% mutate(norm="medianquotientnorm"))
#}
#
##
## quantileNorm
##
#library(preprocessCore)
#tmp1 <- quantileNorm(mw) %>%
#    select(-normalization)
#
#tmp2 <- quantilenorm(mpiu, transpose=TRUE) %>%
#    spread(glycan, value)
#tmp2 <- data.frame(tmp2)
#tmp2 <- tmp2[ , c(2,1,3:ncol(tmp2))]
#
#print(all.equal(tmp1, tmp2))
#if(all.equal(tmp1, tmp2)){
#    mpiunorm <- bind_rows(mpiunorm,
#                          tmp2 %>% mutate(norm="quantilenormt"))
#}
#
#mpiunorm <- mpiunorm %>%
#    gather(glycan, value, contains("GP")) %>%
#    arrange(norm, Plate, gid, glycan)
#mpiunorm <- data.frame(mpiunorm)
#save(mpiunorm, file="./mpiunorm.RData")

#if(.Machine$sizeof.pointer==8){
#
#    data(mpiu)
#    data(mpiunorm)
#
#
#    tanormdf <- mpiunorm %>%
#        filter(norm=="tanorm") %>%
#        select(-norm)
#
#    refpeaknormdf <- mpiunorm %>%
#        filter(norm=="refpeaknorm") %>%
#        select(-norm)
#
#    mediannormdf <- mpiunorm %>%
#        filter(norm=="mediannorm") %>%
#        select(-norm)
#
#    medianquotientnormdf <- mpiunorm %>%
#        filter(norm=="medianquotientnorm") %>%
#        select(-norm)
#
#    quantilenormtdf <- mpiunorm %>%
#        filter(norm=="quantilenormt") %>%
#        select(-norm)
#
#    test_that("tanorm", {
#      expect_equal(tanorm(mpiu), tanormdf)
#    })
#
#
#    test_that("refpeaknorm", {
#      expect_equal(refpeaknorm(mpiu), refpeaknormdf)
#    })
#
#
#    test_that("mediannorm", {
#      expect_equal(mediannorm(mpiu), mediannormdf)
#    })
#
#
#    test_that("medianquotientnorm", {
#      expect_equal(medianquotientnorm(mpiu), medianquotientnormdf)
#    })


#    if(requireNamespace("preprocessCore", quietly=TRUE)){
#        tmp <- quantilenorm(mpiu, transpose=TRUE)
#        # reorder columns
#        tmp <- tmp[, names(quantilenormtdf)] %>%
#            arrange(Plate, gid, glycan)
#
#        test_that("quantilenormt", {
#          expect_equal(tmp, quantilenormtdf)
#        })
#    }
#}