Welcome.

Kingfisher is a fast and flexible program for procurement of sequence files (and their metadata annotations) from public data sources, including the European Nucleotide Archive (ENA), NCBI SRA, Amazon AWS and Google Cloud. It's input is one or more "Run" accessions e.g. DRR001970, or a BioProject accessions e.g. PRJNA621514 or SRP260223.

It has two main modes - the get mode downloads sequence data and the annotate mode downloads metadata.

In the get subcommand, kingfisher downloads data from a series of redundant sources, which it attempts in order until one works. The downloaded data is then converted to an output SRA / FASTQ / FASTA / GZIP file format as required. Both download and extraction phases can be quicker than using the NCBI's SRA toolkit. In particular, downloading from ENA means that FASTQ files are downloaded directly, so there is no need for the extraction step.

If you are looking to download GenBank/RefSeq data instead, then you might try NCBI datasets or ncbi-acc-download.

In the annotation subcommand, metadata about the run(s) is downloaded from NCBI and output in one of several formats e.g. human-readable, CSV, TSV, JSON, feather or parquet. By default, a small amount of metadata is downloaded - more details can be output using the --all-columns flag.

Installation

Installation through bioconda

Kingfisher can be installed in the usual way through conda/bioconda e.g.

conda create -n kingfisher -c conda-forge -c bioconda kingfisher
conda activate kingfisher
kingfisher get -r SRR12118866 -m ena-ftp

Optionally, to use the ena-ascp method, an Aspera connect client is also required. See https://www.ibm.com/aspera/connect/ or https://www.biostars.org/p/325010/.

Installation through DockerHub

A docker image generated from the conda package is available on DockerHub. After installing Docker, run the following, replacing [RELEASE_TAG] with a tag from https://hub.docker.com/r/wwood/kingfisher/tags:

docker pull wwood/kingfisher:[RELEASE_TAG]

To download data into the current directory:

docker run -v `pwd`:/data wwood/kingfisher:[RELEASE_TAG] get -r SRR12118866 -m ena-ftp

Conda-based development environment

Kingfisher can be installed by installing its conda dependencies as follows. Then it can be run from the source directory:

git clone https://github.com/wwood/kingfisher-download
cd kingfisher-download
conda env create -n kingfisher -f kingfisher.yml
conda activate kingfisher
cd bin
export PATH=$PWD:$PATH
kingfisher -h

Optionally, to use the ena-ascp method, an Aspera connect client is also required. See https://www.ibm.com/aspera/connect/ or https://www.biostars.org/p/325010/.

@MakeTheBrainHappy implemented an example installation (and usage) guide in Google Colaboratory and Jupyter Notebook.

Usage

For all modes, a full run-down of the functionality is available using the --full-help flag e.g. kingfisher get --full-help or kingfisher annotate --full-help.

'get' mode: Download and optionally convert sequence data

$ kingfisher get -r ERR1739691 -m ena-ascp aws-http prefetch

This will download .fastq.gz files of the run ERR1739691 from the ENA, or failing that, downloads an .sra file from the Amazon AWA Open Data Program and then converts to FASTQ, or failing that use NCBI prefetch to download and convert that to FASTQ. Kingfisher will do the least effort to convert a downloaded file into one of the formats specified in --output-format-possibilities which is fastq fastq.gz by default.

Output files are put into the current working directory. There are many options for output formats, different download methods etc. Check 'Method details' below and the full help for more details.

'extract' mode: Convert sequence data from .sra format

$ kingfisher extract --sra ERR1739691.sra -t 16 -f fastq.gz

This will extract the file ERR1739691.sra using 16 threads and convert it to fastq.gz file(s). Since this run has paired sequencing data in it, it will create two files ERR1739691_1.fastq.gz and ERR1739691_2.fastq.gz.

'annotate' mode: Get a table of metadata

$ kingfisher annotate -r ERR1739691
run        | bioproject | Gbp   | library_strategy | library_selection | model               | sample_name | taxon_name
---------- | ---------- | ----- | ---------------- | ----------------- | ------------------- | ----------- | ----------
ERR1739691 | PRJEB15706 | 2.382 | WGS              | RANDOM            | Illumina HiSeq 2500 | MM1_1       | metagenome

A fuller set of information is available with --all-columns and the table can also be output in CSV, TSV, JSON, feather or parquet formats using the -f flag.

Method details

In get mode, there are several ways to procure the data:

methoddescription
ena-ascpDownload .fastq.gz files from ENA using Aspera, which can then be further converted. This is the fastest method since no fasterq-dump is required.
ena-ftpDownload .fastq.gz files from ENA using curl, which can then be further converted. This is relatively fast since no fasterq-dump is required.
prefetchDownload .SRA file using NCBI's prefetch from sra-tools, which is then extracted with fasterq-dump.
aws-httpDownload .SRA file from AWS Open Data Program using aria2c with multiple connection threads, which is then extracted with fasterq-dump.
aws-cpDownload .SRA file from AWS using aws s3 cp, which is then extracted with fasterq-dump. Does not usually require payment or an AWS account.
gcp-cpDownload .SRA file from Google Cloud gsutil, which is then extracted with fasterq-dump. Requires payment and a Google Cloud account.

The ena-ascp method of this tool was built based on the very helpful bio-stars thread written by @ATpoint. To find run identifiers to be used as input to kingfisher, you might find the SRA explorer site helpful.

Near parity with ena-fast-download

Ena-fast-download was the original name for this tool. To imitate that tool's functionality:

kingfisher get -r ERR1739691 -m ena-ascp

FAQ

ascp: not found

If you see this error /bin/sh: 1: ascp: not found as below:

$ kingfisher get -r ERR3357550 -m ena-ascp
05/04/2021 05:13:45 AM INFO: Attempting download method ena-ascp ..
05/04/2021 05:13:45 AM INFO: Using aspera ssh key file: $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh
05/04/2021 05:13:45 AM INFO: Querying ENA for FTP paths for ERR3357550..
05/04/2021 05:13:49 AM INFO: Downloading 2 FTP read set(s): ftp.sra.ebi.ac.uk/vol1/fastq/ERR335/000/ERR3357550/ERR3357550_1.fastq.gz, ftp.sra.ebi.ac.uk/vol1/fastq/ERR335/000/ERR3357550/ERR3357550_2.fastq.gz
05/04/2021 05:13:49 AM INFO: Running command: ascp -T -l 300m -P33001  -i $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh era-fasp@fasp.sra.ebi.ac.uk:/vol1/fastq/ERR335/000/ERR3357550/ERR3357550_1.fastq.gz .
/bin/sh: 1: ascp: not found
05/04/2021 05:13:49 AM WARNING: Error downloading from ENA with ASCP: Command 'ascp -T -l 300m -P33001  -i $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh era-fasp@fasp.sra.ebi.ac.uk:/vol1/fastq/ERR335/000/ERR3357550/ERR3357550_1.fastq.gz .' returned non-zero exit status 127.
05/04/2021 05:13:49 AM WARNING: Method ena-ascp failed
Traceback (most recent call last):
  File "./bin/kingfisher", line 330, in <module>
    raise Exception("No more specified download methods, cannot continue")
Exception: No more specified download methods, cannot continue

then you have not installed the Aspera client correctly. See the installation section of this document.

Failed to authenticate with ascp

This error manifests like this

$ kingfisher get -r SRR5005053 -m ena-ascp
10/26/2021 11:24:21 PM INFO: Kingfisher v0.0.1-dev
10/26/2021 11:24:21 PM INFO: Attempting download method ena-ascp ..
10/26/2021 11:24:21 PM INFO: Using aspera ssh key file: $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh
10/26/2021 11:24:21 PM INFO: Querying ENA for FTP paths for SRR5005053..
10/26/2021 11:24:33 PM INFO: Downloading 2 FTP read set(s): ftp.sra.ebi.ac.uk/vol1/fastq/SRR500/003/SRR5005053/SRR5005053_1.fastq.gz, ftp.sra.ebi.ac.uk/vol1/fastq/SRR500/003/SRR5005053/SRR5005053_2.fastq.gz
10/26/2021 11:24:33 PM INFO: Running command: ascp -T -l 300m -P33001 -k 2 -i $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh era-fasp@fasp.sra.ebi.ac.uk:/vol1/fastq/SRR500/003/SRR5005053/SRR5005053_1.fastq.gz .
10/26/2021 11:24:45 PM WARNING: Error downloading from ENA with ASCP: Command ascp -T -l 300m -P33001 -k 2 -i $HOME/.aspera/connect/etc/asperaweb_id_dsa.openssh era-fasp@fasp.sra.ebi.ac.uk:/vol1/fastq/SRR500/003/SRR5005053/SRR5005053_1.fastq.gz . returned non-zero exit status 1.
STDERR was: b'ascp: failed to authenticate, exiting.\n'STDOUT was: b'\r\nSession Stop  (Error: failed to authenticate)\n'
10/26/2021 11:24:45 PM WARNING: Method ena-ascp failed
Traceback (most recent call last):
  File "/analysis2/software/kingfisher-download/bin/kingfisher", line 280, in <module>
    main()
  File "/analysis2/software/kingfisher-download/bin/kingfisher", line 232, in main
    kingfisher.download_and_extract(
  File "/analysis2/software/kingfisher-download/bin/../kingfisher/__init__.py", line 37, in download_and_extract
    download_and_extract_one_run(run, **kwargs)
  File "/analysis2/software/kingfisher-download/bin/../kingfisher/__init__.py", line 276, in download_and_extract_one_run
    raise Exception("No more specified download methods, cannot continue")
Exception: No more specified download methods, cannot continue

This could be caused by (1) your being on a network that interferes with ascp operation, or (2) a temporary downtime at ENA. You may try moving to a different network or following the instructions at the Aspera support or checking the log files to see if that helps diagnose the error. See this issue for more information about a specific WLAN that was causing an issue.

API rate limit exceeded

Using kingfisher annotate repeatedly and in parallel can mean that the default allowable rate of requests (3 per second) is exceeded. To get around this error, you can generate an NCBI API key and then set the environment variable NCBI_API_KEY before running kingfisher. This key will then be used in all requests.

Citation

Woodcroft, B. J., Cunningham, M., Gans, J. D., Bolduc, B. B., & Hodgkins, S. B. (2023). Kingfisher: A utility for procurement of public sequencing data. [Computer software]. DOI: 10.5281/zenodo.10525085.

DOI

License

Copyright Ben Woodcroft 2019-2023. Licensed under GPL3+. See LICENSE.txt.

The source code is available at https://github.com/wwood/kingfisher-download.

Powered by Doctave