sradb RNA-seq samps by species

by_species.R

library('SRAdb')

# If you have SRAmetadb already, set this directory, or setwd appropriately
# to save yourself a large download
sqlfile <- file.path('.', 'SRAmetadb.sqlite')
if(!file.exists('SRAmetadb.sqlite')) sqlfile <<- getSRAdbFile()
sra_con <- dbConnect(SQLite(),sqlfile)
q <- function(x) { dbGetQuery(sra_con, x) }

species_to_tax_id = list(
  'arabidopsis_thaliana'=3702,     # 4. arabidopsis
  'bos_taurus'=9913,               # 9: cow
  'caenorhabditis_elegans'=6239,   # >10: roundworm
  'danio_rerio'=7955,              # 3: zebrafish
  'drosophila_melanogaster'=7227,  # 5: fruitfly
  'homo_sapiens'=9606,             # 2: human
  'mus_musculus'=10090,            # 1: mouse
  'ovis_aries'=9940,               # 10: sheep
  'rattus_norvegicus'=10116,       # 7: rat
  'saccharomyces_cerevisiae'=4932, # 6: yeast
  'zea_mays'=4577)                 # 8: corn

# > sort(table(taxids),decreasing=TRUE)[1:10]
# taxids
#  10090   9606   7955   3702   7227   4932  10116   4577   9913   9940 
# 195878 156394  23590  17296  17006   7997   7098   7018   5173   5018 

species_table <- function(taxid) {
  q(paste(
    'SELECT * FROM sra, study, submission, fastq, run',
    'WHERE sra.platform = "ILLUMINA"',
    '  AND sra.library_strategy = "RNA-Seq"',
    '  AND sra.library_source = "TRANSCRIPTOMIC"',
    '  AND sra.submission_accession = submission.submission_accession',
    '  AND sra.study_accession = study.study_accession',
    '  AND sra.run_accession = fastq.run_accession',
    '  AND sra.run_accession = run.run_accession',
    '  AND fastq.FASTQ_FILES > 0',
    sprintf('AND sra.taxon_id = %d', taxid)))
}
runs_by_species <- lapply(species_to_tax_id, species_table)