data-raw/midasok.R

source("data-raw/helper-functions.R")
require("data.table")

data("mfg_functions")
metadata <- fread("data-raw/midas_onlinekontrol/merged_data/OnlineKontrol/merged_metadata.csv")

#this is hopefully FALSE, otherwise manually check which samples to remove:
if(any(duplicated(metadata[[1]])))
  stop("Duplicated sample IDs")

# filter/rename a few columns
metadata <- metadata[
  ,
  .(
    SampleID = Barcode,
    Plant = SampleSite,
    Date,
    Line,
    Tank,
    SampleContent
  )
]

# rename some values
metadata[grepl("activated", tolower(SampleContent)), SampleContent := "AS"]
metadata[grepl("digester", tolower(SampleContent)), SampleContent := "AD"]

midasok <- amp_load(
  otutable = "data-raw/midas_onlinekontrol/merged_data/OnlineKontrol/merged_otutable.csv.gz",
  metadata = metadata
)

#this step uses an awful lot of memory
midasok <- filter_otus(midasok, 0.1)

midasok$metadata$Date <- lubridate::ymd(midasok$metadata$Date)
midasok$metadata <- fix_metadata(midasok$metadata)
midasok <- genusfunctions(
  midasok,
  function_data = mfg_functions
)

usethis::use_data(midasok, overwrite = TRUE)
#midasok_PeriodAvg <- periodAvg(midasok$metadata)
#usethis::use_data(midasok_PeriodAvg, overwrite = TRUE)
cmc-aau/midasdashboard documentation built on Jan. 29, 2024, 1:42 p.m.