Once we have collected all our SRA numbers in our text file we then need to download the files to our local system or the cluster. We use the SRA tools tool called prefetch to download the SRA files.

Firstly load the sra2peak virtual environment, we need to remain in this environment until the peak calling step.

source activate sra2peak

Then qquickly check if prefetch is installed correctly and in you current path type:

prefetch 

If it produces a usage statement we’re good to go if not, the package is either not on your system or not in your path. See Installing required packages page.

To download the SRA files we write a short for loop.

for file in $(<sra_list.txt)
do
    prefetch ${file}
done

This effectively loops through the SRA files in the list we generated and runs one instance of prefetch on each file separately.

One slight annoyance with prefetch is there is not a simple way to sepcify the output folder for the SRA files they go to the following location /home/[USER]/ncbi/public/sra/. This can be changed but is a bit of a faff and beyond the scope of this tutorial. If you really want to change it see here, but we can work around this in the next command.

Quick check to see if SRA files are actually type specifed in repo

On occasion files uploaded to the SRA repo will be annotated incorrectly, i.e. the will say the are single ended when in fact the are paired ended. It may not be immediately obvious that this has occured so you can do a quick sanity check by running this short bash script.

~~bash #!bin/bash

for file in $(<sra_list.txt) do

srr=”SRR2920543.sra” numLines=$(fastq-dump -X 1 -Z –split-spot $srr | wc -l) if [ $numLines -eq 4 ] then echo “$srr is single-end” else echo “$srr is paired-end” fi

done ~~

This will run one read (takes milliseconds) and quickly tells you if the SRA file is single or paired ended. See here for more info/options.

Move on to SRA to fastq, or back to Get SRA file information.