sra-downloader
A script for batch-downloading and automatic compression of data from NCBI Sequence Read Archive. Built on SRA-Toolkit, which ensures high throughput during the download.
Features:
- Batch downloading reads for whole studies
- Downloading SRAs using either accession IDs or NCBI generated files
- Organizing sequences by projects that they come from
- Detecting which runs have been already downloaded
Requirements
- python >= 3.7
- sra-toolkit >= 2.9.6
- pigz
Instalation
0. Docker
- Run
docker run wwydmanski/sra-downloader -h
1. Conda (recommended, also downloads sra-toolkit)
- Run
conda install -c bioconda -c bioinf-mcb sra-downloader
Note: if you don't specify the bioconda channel you will get a dependency error.
2. From PyPi
- Run
pip install sra-downloader
3. From sources
- Download a repo into a folder
- Run
pip install .
Usage
usage: sra-downloader [-h] [--fname FILENAME] [--save-dir SAVE_DIRECTORY] [--uncompressed [UNCOMPRESSED]] [--cores [CORES]] [sra_id [sra_id ...]]
Download SRA data and organize them by projects
positional arguments:
sra_id SRA IDs to download
optional arguments:
-h, --help show this help message and exit
--fname FILENAME CSV file with list of SRAs to download. Header must include `Run` and `BioProject`.
--save-dir SAVE_DIRECTORY
a directory that the files will be saved to. (default: ./downloaded)
--uncompressed [UNCOMPRESSED]
if present, the files will not be compressed. (default: False)
--cores [CORES] Cores used for parallel compression. (default is the number of online processors, or 8 if unknown)
Examples
sra-downloader ERR2177760 --uncompressed
docker run -v $(pwd)/downloads:/downloaded wwydmanski/sra-downloader ERR1551967
sra-downloader --fname SraRunTable.txt --save-dir ./SRAs --cores 4
Sample output
└─── save_folder
├── PRJEB14961
│ ├── ERR1551967.sra_1.fastq.gz # - raw read archived files from SRA
│ ├── ERR1551967.sra_2.fastq.gz
│ └── SraRunTable.txt # - original SraRunTable.txt with useful metadata about samples
└── PRJEB20463
├── ERR2177760.sra_1.fastq.gz
├── ERR2177760.sra_2.fastq.gz
├── absent.txt # - entries that were unaccessible due to various reasons
└── SraRunTable.txt
