For a file of SRA run IDs (ERR457868 etc.), ask the Sequence Read Archive for the associated BioSample names (SAMEA2399445 etc.)

SRA_Runs_to_BioSample.R

library(rentrez)
library(assertthat)
library(readr)
search_ind <- function(term){
  # get the IDs for a run ID
  # ERR457868 searched, returns 1011219
  results <- entrez_search(db="sra", term=term)$ids
  assert_that(length(results) == 1)
  results
}

search_links <- function(term) {
  # get the Links in the SRA for a specific run's ID
  # then pull out only the biosample from there
  results <- entrez_link(dbfrom='sra', id=term, db='biosample')$links$sra_biosample
  # there should be only one biosample per ID
  assert_that(length(results) == 1)
  results
}

search_summary <- function(term) {
  # Get the summary for a BioSample, then return only the accession
  summary <- entrez_summary(db='biosample', id=term)$accession
  summary
}

names <- read.table('./Names_only.txt', head=F)
#ERR457868 ERR475358 ERR475359 ERR475360 ERR475361 ERR479604 ....
ids <- sapply(names$V1, search_ind, USE.NAMES=F)
#[1] "1011219" "1533032" "984971"  "984969"  "984970"  "1533033"....
links <- sapply(ids, search_links, USE.NAMES=F)
# "3087115" "3769628" "3031276" "3031277" "3031278" "3769630"
accession_ids <- sapply(links, search_summary, USE.NAMES=F)
# [1] "SAMEA2399445" "SAMEA2445339" "SAMEA2729910" "SAMEA2445340" "SAMEA2445341" "SAMEA2467095"
names$V2 <- accession_ids
write_csv(names, 'IDs_to_Samples.csv')