inst/scripts/make-metadata.R

In MahShaaban/curatedAdipoArray: A Curated Microarrays Dataset of MDI-induced Differentiated Adipocytes (3T3-L1) Under Genetic and Pharmacological Perturbations

# load libraries --------------------------------------------------------------
library(tidyverse)
library(SummarizedExperiment)

# get list of clean data files ------------------------------------------------
fls <- list.files('.', pattern = 'GSE*')
names(fls) <- fls

# extract phenotype data from each file ---------------------------------------
md1 <- map(fls, function(x) {
    gse <- read_rds(x)
    as.data.frame(colData(gse))
  }) %>%
  bind_rows(.id = 'path')

# make metadata table for clean data files ------------------------------------
md1 %>%
  group_by(series_id) %>%
  summarise(
    Title = paste0(
      "A Clean Expression Matrix of the GEO", unique(series_id), " Dataset."
    ),
    Description = paste0(
      "MDI-induced 3T3-L1 pre-adipocytes with ",
      paste0(unique(treatment_type), collapse = '/'),
      " treatment for ",
      paste0(unique(treatment_target), collapse = '/')
    ),
    BiocVersion = "3.11",
    Genome = 'mm10',
    SourceType = "GSEMatrix",
    SourceUrl = "https://github.com/MahShaaban/curatedAdipoArray",
    SourceVersion = "October 10 2019",
    Species = "Mus musculus",
    TaxonomyId = "10090",
    Coordinate_1_based = TRUE,
    DataProvider = "GEO",
    Maintainer = "Mahmoud Ahmed <mahmoud.s.fahmy@students.kasralainy.edu.eg>",
    RDataClass = "SummarizedExperiment",
    DispatchClass = "Rds",
    RDataPath = paste0('curatedAdipoArray/', unique(path)),
    Tags = "",
    Notes = ""
  ) %>%
  select(-series_id) %>%
  write_csv('inst/extdata/metadata.csv')

# extract phenotype data from processed files ---------------------------------
fls <- list.files('.', pattern = '*_perturbations*')

# make metadata table for clean processed files -------------------------------
data.frame(path = fls) %>%
  mutate(
    type = str_split(path, '\\_|\\.', simplify = TRUE)[, 1],
    batch = ifelse(grepl('2', path), '(processed)', ''),
    Title = paste(
      "A Curated Microarrays Dataset", batch,"of MDI-induced",
      "Differentiated Adipocytes (3T3-L1) Under", type,"Perturbations"
    ),
    Description = paste(
      "A curated dataset", batch, "of Microarrays",
      "samples. The samples are MDI-induced pre-adipocytes (3T3-L1) at",
      "different time points/stage of differentiation under different types",
      "of", type,"perturbations."
    ),
    BiocVersion = "3.11",
    Genome = "mm10",
    SourceType = "GSEMatrix",
    SourceUrl = "https://github.com/MahShaaban/curatedAdipoChIP",
    SourceVersion = "October 10 2019",
    Species = "Mus musculus",
    TaxonomyId = "10090",
    Coordinate_1_based = TRUE,
    DataProvider = "GEO",
    Maintainer = "Mahmoud Ahmed <mahmoud.s.fahmy@students.kasralainy.edu.eg>",
    RDataClass = "SummarizedExperiment",
    DispatchClass = "Rds",
    RDataPath = paste0('curatedAdipoArray/', path),
    Tags = "",
    Notes = ""
  ) %>%
  select(-path, -type, -batch) %>%
  write_csv('inst/extdata/metadata.csv', append = TRUE)