iSeq: Anintegrated tool to fetch publicSequencing data

Description

Important

To use iSeq, Your system must beconnected to the network andsupport FTP, HTTP, and HTTPS protocols.

Update Notes:

2025.03.14

Features

Installation

1.iSeq can be installed by conda easily

conda install bioconda::iseq

2. The latest version of iSeq can also be installed from source, seeINSTALL

# Use the following command to check whether dependent software is installediseq --version

Example (See more)

iseq -i PRJNA211801

iseq -i SRR_Acc_List.txt -a -g

Usage (中文教程✨)

$ iseq --helpUsage:  iseq -i accession [options]Required option:  -i, --input     [text|file]   Single accession or a file containing multiple accessions.                                Note: Only one accession per line in the fileOptional options:  -m, --metadata                Skip the sequencing data downloads and only fetch the metadata for the accession.  -g, --gzip                    Download FASTQ files in gzip format directly (*.fastq.gz).                                Note: if *.fastq.gz files are not available, SRA files will be downloaded and converted to *.fastq.gz files.  -q, --fastq                   Convert SRA files to FASTQ format.  -t, --threads   int           The number of threads to use for converting SRA to FASTQ files or compressing FASTQ files (default: 8).  -e, --merge     [ex|sa|st]    Merge multiple fastq files into one fastq file for each Experiment, Sample or Study.  -d, --database  [ena|sra]     Specify the database to download SRA sequencing data (default: ena).                                Note: new SRA files may not be available in the ENA database, even if you specify "ena".  -p, --parallel  int           Download sequencing data in parallel, the number of connections needs to be specified, such as -p 10.                                Note: breakpoint continuation cannot be shared between different numbers of connections.  -a, --aspera                  Use Aspera to download sequencing data, only support GSA/ENA database.  -s, --speed     int           Download speed limit (MB/s) (default: 1000 MB/s).  -o, --output    text          The output directory. If not exists, it will be created (default: current directory).  -h, --help                    Show the help information.  -v, --version                 Show the script version.

1.-i,--input

iseq -i PRJNA211801

2.-m,--metadata

iseq -i PRJNA211801 -miseq -i CRR343031 -m

Note

Note 1: If the retrieved accession is in theSRA/ENA/DDBJ/GEO databases,iSeq will first search in the ENA database. If sample information can be retrieved, it will download metadata inTSV format via theENA API, typically containing 191 columns. However, some recently released data in the SRA database may not be promptly synchronized to the ENA database. Therefore, if metadata cannot be obtained from the ENA database,iSeq will directly download metadata inCSV format via theSRA Database Backend, typically containing 30 columns. To maintain consistency with the TSV format, it will be converted to TSV format usingsed -i 's/,/\t/g'. However, if a single field contains a comma, it may cause column disorder. Ultimately, you will obtain sample information named${accession}.metadata.tsv.

Note

Note 2: If the retrieved accession is in theGSA database,iSeq will obtain sample information via GSA'sgetRunInfo interface, downloading metadata inCSV format, typically containing 25 columns. The metadata obtained above will be saved as${accession}.metadata.csv. To supplement more detailed metadata information, iSeq will automatically obtain metadata information for the Project to which the accession belongs via GSA'sexportExcelFile interface, downloading metadata inXLSX format, typically with 3 sheets:Sample,Experiment,Run. The final metadata information will be saved as${accession}.metadata.xlsx. In summary, you will ultimately obtain sample information named${accession}.metadata.csv andCRA*.metadata.xlsx.

3.-g,--gzip

iseq -i SRR1178105 -g

Tip

parallel-fastq-dump can also convert SRA to gzip-compressed FASTQ files, typically2-3 times faster thanfasterq-dump + pigz. However, consideringIO limitations,iSeq currently does not supportparallel-fastq-dump.

4.-q,--fastq

iseq -i SRR1178105 -q

Note

Note1:-q is particularly useful for downloadingsingle-cell data, especially for scATAC-Seq data, as it can effectively decompose the files into four parts:I1,R1,R2,R3. However, if FASTQ files are directly downloaded via the-g parameter, onlyR1 andR3 files will be obtained (e.g.,SRR13450125), which may cause issues during subsequent data analysis.

Note

Note 2: When-q and-g are used together, the SRA file will first be downloaded, then converted toFASTQ files using thefasterq-dump tool, and finally compressed into gzip format usingpigz. It does not directly downloadFASTQ files in gzip format, which is very useful for obtaining comprehensive single-cell data.

5.-t,--threads

iseq -i SRR1178105 -q -t 10

6.-e,--merge

iseq -i SRX003906 -g -e ex

Note

Note 1: If the accession is aRun ID, the-e parameter cannot be used (see below). Currently,iSeq supports mergingboth gzip-compressed and uncompressed FASTQ files, but does not support merging files such asBAM files and tar.gz files.

Note

Note 2: Normally, when an Experiment contains only one Run, identical Runs should have thesame prefix. For example,SRR52991314_1.fq.gz andSRR52991314_2.fq.gz have the same prefixSRR52991314. In this case,iSeq will directlyrename them toSRX*_1.fastq.gz andSRX*_2.fastq.gz. However, there are exceptions, such as inCRX006713 where a RunCRR007192 contains files with different prefixes. In such cases,iSeq willrename them asSRX*_original_filename, for example, they will be renamed asCRX006713_CRD015671.gz andCRX006713_CRD015672.gz.

7.-d,--database

iseq -i SRR1178105 -d sra

Note

Note: If the corresponding SRA file is not found in theENA database, even if the-d ena parameter is specified,iSeq will still automatically switch to downloading from theSRA database.

8.-p,--parallel

iseq -i PRJNA211801 -p 10

Note

Note 1: Theresumable download feature of multi-threaded downloading is only effective within thesame thread. That is, if the-p 10 parameter is used for the first download, it must also be used for the second download to enable resumable download.

Note

Note 2: As mentioned,iSeq will maintain 10 connections throughout the download process. Therefore, you will see multiple occurrences of the sameConnection * finished popping up during the download process. This is because some connections are released immediately after completing the download and then new connections are established for downloading.

9.-a,--aspera

iseq -i PRJNA211801 -a -g

Note

Note 1: When accessing theGSA database, if download links fromHuawei Cloud are available,iSeq will prioritize downloading through Huawei Cloud, even if the-a parameter is used. This is because Huawei Cloud offers faster and more stable download speeds. Therefore, when downloading GSA data, it'srecommended to use the-a parameter. This way, if access to Huawei Cloud is unavailable, downloading through the Aspera channel is still relatively fast. Otherwise, you'll have to resort to downloading viawget oraxel, which are slower methods.

Note

Note 2: SinceAspera requires a key file,iSeq willautomatically search for the key file in theconda environment or the~/.aspera directory. If the key file is not found, downloading will not be possible.

10.-o,--output

11.-s,--speed

Output

Inspired

Contributing

License