get FASTQ files from NCBI

To understand signaling pathways, we intend to find networks using WGCNA. For this purpose, we require RNA-seq data from cancer samples at NCBI, and this information is available in the Gene Expression Omnibus database at NCBI. “A GEO Series (GSExxx) is an original submitter-supplied record that summarizes a study” — NCBI.

The first step is to visit the SRA Run Selector from the Accession display of the Series. From the SRA Run Selector , download the metadata as SraRunTable. Next, load the SraRunTable of all the datasets in R. Optional- choose 50 random samples of the Run column of SraRunTable and save as .csv files, or choose all the entries from the Run column. For obtaining the FASTQ files, prefetch and fasterq from the NCBI SRA toolkit is used. An important issue to consider is disk space. Around 17 times the size of the samples in the accession is needed for running the files. The process is well explained in https://github.com/ncbi/sra-tools/wiki/08.-prefetch-and-fasterq-dump. prefetch will create a directory and download the .sra format run file in the directory. For all the accessions in the .csv file, a simple BASH shell script can handle all the files. Then, one can run the fasterq-dump as a BASH shell script to convert all .sra files to FASTQ format.

[embed]https://gist.github.com/me-nobody/b1679da1ac28dc41c6a0a62cc68d1c61