Download the data sets identified in the linked ticket into the "data processing" directory on the cluster.

Instructions:

1) For each "SRP" accession (the identifier that uniquely identifies a data set), create a folder with the same name as the dataset. For example, if the dataset accession is SRP123456, make a folder named SRP123456. Make sure the folder is group-writeable so that other people in the group can add files to it. (By default, it will likely already be group-executable and group-readable, but please check it just in case.)

2) Use the "run selector" tool to make a meta-data file listing all the "runs" (SRR id's) plus library names, sample names, etc. for each "SRP" accession. Add it to the data set's folder.

3) Download fastq files and save them to a subfolder named "fastq" within each "SRP" folder.

4) Check that the data downloaded correctly. Talk with Molly and Rob Reid about how best to do that. Also, do some research (google, biostars, etc) for fast and easy ways to check that the download process did not fail. Another possibility: run fastQC on the data and compare the output to the meta-data file from the run selector tool? If you notice any problems, fix them.

5) Compress all the fastq files using gzip.