Download public data

RNA-seq data can be downloaded from GEO, NCBI’s repository of transcriptomics data. However, GEO only contains the counts, not the reads. The reads are stored in another NCBI database: the SRA database.

There are 2 ways to download data from NCBI. They will be explained in the videos below.

Download via web browser

In the video we download a sample from this bulk RNA-seq data set.

When you download FASTQ files from SRA via your browser they are often in SRA Lite format. That means that each base is given a quality score of either 30 (represented by ?) or 3 (represented by $). This makes the FASTQ files a lot smaller and faster to download.

Download via SRA Toolkit

If you want FASTQ files with original quality scores you need to do the download using the SRA Toolkit.

For this you need the SRA Run ID of the sample (= the name of the sample in the SRA database). This number starts with the letters SRR and can be found in the table of the SRA record of the sample.

The code for downloading a sample with Run ID SRR34669421 is as follows:

prefetch SRR34669421
fasterq-dump SRR34669421