get_SRA_file_URL_for_lib_GEO_accession.sh

# Problem: You have a NCBI GEO accession and would like to get the URL of the SRA file that contains the sequencing data.
# The sed command that removes the last characer of the string is essential as there is a invisible character that messes up the
# downstream steps otherwise.

GEO_ACCESSION="GSM1655353" # set you GEO accession here
SRA_FTP_URL=$(curl "http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=${GEO_ACCESSION}&targ=self&form=text&view=brief" 2>/dev/null | grep ftp-trace.ncbi.nlm.nih.gov | cut -c 32-| sed 's/.$//')
FTP_SUB_FOLDER=$(ncftpls ${SRA_FTP_URL}/)
SRA_FILE=$(ncftpls ${SRA_FTP_URL}/${FTP_SUB_FOLDER}/)
echo $GEO_ACCESSION ${SRA_FTP_URL}/${FTP_SUB_FOLDER}/${SRA_FILE}