data-raw/lipodomics.R

In Mortendall/Lipodomics: What the Package Does (One Line, Title Case)

## code to prepare `lipodomics` dataset goes here



count_matrix <- count_file_loader("lipid_NR.xlsx")
setup <- load_metadata("setup.xlsx")

setup <- setup %>%
  unite(Group, Genotype, NR, sep = "_", remove = F)


#Might not be beneficial to remove low counts as data are percentile
#count_matrix <- select_sufficient_counts(count_matrix,setup,2)




#as data is normalized pr info from Mesut, they should be directly comparable
#dir.create(here::here("data/figures"), showWarnings = F)
#Quality_control_plots(count_matrix_group,setup)
#rename folder so it doesn't overwrite
#Quality_control_plots(count_matrix_sub,setup)

#based on QC, 330 is excluded (was steatotic)

setup <- setup %>%
  filter(!ID=="330")
colnames(count_matrix) <- as.integer(colnames(count_matrix))

count_matrix <- count_matrix %>%
  dplyr::select(-"330")

all(setup$ID==colnames(count_matrix))

count_matrix_groups <- count_matrix[217:238,]
count_matrix_sub <- count_matrix[1:216,]


#sum(count_matrix_groups/26)
#sum(count_matrix_sub/26)
#now data are organized as 100% pr column


test_group <- multiple_t_test_lipids(count_matrix_groups, setup)
test_sub <- multiple_t_test_lipids(count_matrix_sub, setup)

#heatmap<-heatmap(count_matrix,setup)

#Repeat for analysis without TG and cholesterol


count_matrix_lim <- count_file_loader("lipid_NR_no_TAG_chol.xlsx")
count_matrix_lim <- count_matrix_lim %>%
  dplyr::select(-"330")

all(setup$ID==colnames(count_matrix))

count_matrix_lim_groups <- count_matrix_lim[206:225,]
count_matrix_lim_sub <- count_matrix_lim[1:205,]

test_lim_group <- multiple_t_test_lipids(count_matrix_lim_groups, setup)
test_lim_sub <- multiple_t_test_lipids(count_matrix_lim_sub, setup)