Downloading data from dbGaP (IN PROGRESS)

The scripts for downloading SRA files from dbGaP are in this repository. Some scripts have been adapted from Hurwitz Lab for use on Phoenix HPC.

Some archives on dbGaP have elected to store and distribute their data from cloud providers in their originally submitted format. These files are not available via the SRA toolkit and require an Amazon or Google cloud serves to access these reads.

Once access to files has been granted, please select the run selector option from the Actions tab. You can check the format of the files as well as the Datastore provider using the DATASTORE filetype and DATASTORE region columns.

If the DATASTORE region is ncbi, then these files can be downloaded using the SRA tool-kit

Retrive files from dbGaP using SRA-toolkit

For more instructions, visit this ncbi document. Scripts in this repository are formatted for Phoenix HPC.

1. Download the repository key and move it to $FASTDIR on phoenix. After following the instructiions from the ncbi guide, download the cart file with your selected files and move to $FASTDIR.

2. Run the get_dbgap.sh script. This script will create the main repository that allows you to download from the cart file. This script also contains the prefetch command which works with bulk download from cart files.

Some helpful documents

• Fastq-dump by Edwards lab