get_group_averages()
is a wrapper function to return group averages
for the filtered peak table.
Value
a data.table
reporting the average and relative standard
deviation across biological groups and technical replicates within
each group.
Examples
data <- import_data(
example_path("coculture_peak_table.csv"),
example_path("metadata.csv"),
format = "Progenesis"
)
data_filter <- filter_group(data, group_to_remove = "Blanks")
#> ℹ Parsing 1303 peaks based on the sample group: Blanks.
#> ℹ Argument remove_ions is: TRUE.Removing peaks from Blanks.
#> ✔ 796 ions failed the Blanks filter, 507 ions remain.
group_averages <- get_group_averages(data_filter)
head(group_averages)
#> Compound Biological_Group average BiolRSD Bioln techRSD techn
#> <char> <char> <num> <num> <int> <num> <num>
#> 1: 1000 ANG18 monoculture 0.000 NA 3 NA 3
#> 2: 1000 ANGDT monoculture 3762.133 0.01990561 3 0.01990561 3
#> 3: 1000 Blanks 0.000 NA 3 NA 3
#> 4: 1000 Coculture 0.000 NA 3 NA 3
#> 5: 1000 JC1 monoculture 0.000 NA 3 NA 3
#> 6: 1000 JC28 monoculture 182681.783 0.01643330 3 0.01643330 3