Downloads
Download, install, and get started using the DNAnexus Platform SDK, the DNAnexus upload and download agents, and dxCompiler.
DNAnexus Platform SDK
The DNAnexus Platform SDK - also called dx-toolkit
- includes the dx
command-line client; tools for building and debugging apps; utilities for working with DNA data on the DNAnexus Platform; and Python, Java, C++ and R bindings for working on the DNAnexus Platform.
The SDK is released under the Apache v2.0 license. This SDK does not collect any user identifiable information.
If you plan to rely on the user interface to interact with the Platform, using the command-line interface only to upload files, you likely don't need to download the full SDK. The Upload Agent may be enough to meet your needs.
Installing the Python SDK and Command Line Tools
Before You Begin
Before installing the Python SDK and command line tools, ensure that you are running Python 3.
Installation
The SDK's Python components - dxpy
bindings and command line tools, including dx
- are distributed through PyPI. Use pip3 to download and install the dxpy
package:
If you see a warning similar to the one below, please update your PATH environment variable to include the path where pip3 installed the new commands.
Enabling Tab Completion
If you're using Z shell (zsh), which is included in recent versions of OS X, enable tab completion by entering the following command, or adding it to your .zshrc
:
If you're not using Z shell (zsh), enable tab completion by entering the following command, or add it to your .bashrc
:
Upgrading dxpy
To upgrade the dxpy
package after installing it using pip3, enter the command:
pip3 install --upgrade dxpy
Troubleshooting
If you're using the OS X Xcode command line tools and running Python 3, the psutil dependency may fail to compile. To prevent this from happening, set export ARCHFLAGS="-arch x86_64"
before installing dxpy
.
To avoid character set problems when running the dx
command, set the PYTHONIOENCODING environment variable to UTF-8 as follows:
export PYTHONIOENCODING=UTF-8
On Windows, Windows 10 and later versions are supported.
Building from Source
See the Readme.md
file in the dx-toolkit
Github repository for information on build-time dependencies, Java, C++ and R bindings and additional installation instructions.
Using the SDK
Using a Proxy
To direct dx-toolkit
to a proxy, set the HTTP_PROXY
or HTTPS_PROXY
environment variable to an appropriate value:
export HTTP_PROXY=hostname:port
TLS Settings
By default, dx-toolkit
uses certificates installed on your system to create secure connections. If your system requires an additional TLS certificate, and dx-toolkit
doesn't appear to be using a certificate installed on your system, you have two options. The first, and most secure, is to set the DX_CA_CERT
environment variable to the path of the PEM-encoded TLS certificate file required by your parent organization. The second option is to connect insecurely, avoiding certificate verification altogether, by setting DX_CA_CERT=NOVERIFY
.
When using Windows, environments can be customized in a wide variety of ways. As a result, issues can arise even when using the TLS settings detailed in the previous paragraph. Contact DNAnexus Support if you are using Windows and have trouble connecting through a proxy.
Learn More
The Command Line Quickstart will get you started on accessing the platform via the command-line interface.
The Developer section contains comprehensive information on how to use the SDK to interact with, and write your own applications for the Platform.
Upload Agent
The DNAnexus Upload Agent is a command-line tool for uploading files to the DNAnexus Platform. Using the Upload Agent is a fast and convenient alternative to uploading files via the Platform UI.
Using the Upload Agent requires an authentication token. Please create an authentication token before proceeding.
Features
Fast Parallel Uploads - Data is uploaded using multiple parallel connections for faster upload speeds.
Automatic Compression - Text files (such as FASTA, FASTQ, qseq, and other text files) can be automatically compressed before upload, for greater efficiency.
Automated Uploads - No user interaction is required; the client can be used to automate uploads of entire sequencing runs.
Resumable Uploads - Previously interrupted uploads can be automatically resumed.
Progress Reporting - Users can monitor the progress of individual file uploads, as well as average/instantaneous transfer rates.
Installing the Upload Agent
The Upload Agent is available pre-built for 64-bit Linux, OS X, and Windows.
Installation on 64-bit Linux
Install the Upload Agent on 64-bit Linux as follows:
Installation on OS X
Install the Upload Agent on OS X as follows:
Installation on Windows
To install the Upload Agent on Windows:
Unzip the file using your favorite utility
Run
ua.exe
Using the Upload Agent
See this page for full details on the features and usage of the Upload Agent.
Download Agent
The DNAnexus Download Agent is a command-line tool for downloading large numbers of files from the DNAnexus Platform to a local environment. It is designed to ensure the success of downloads that can take a long time to complete.
Features and Usage
For a detailed description of Download Agent features and usage, please review the README file in the Download Agent Github repository.
Downloads
A list of Download Agent releases is available in Releases section of the Download Agent Github repository. The Download Agent Docker repository provides access to experimental builds corresponding to commits to the master branch.
dxCompiler
dxCompiler is a tool for building a DNAnexus workflows from a workflows written in Workflow Description Language (WDL). It is distributed as a JAR file that can be executed using Java (version 8 or 11).
Installing dxCompiler
To set up and use dxCompiler, follow these steps:
Run
dx login
to log into DNAnexusSelect a dxCompiler execution method:
Using a JAR file
Download the dxCompiler JAR file from the releases section of the dxCompiler Github repository
Install Java SDK 8 or 11
Run dxCompiler using
java -jar dxCompiler.jar compile workflow_file.wdl
Using a Docker container
Build a Docker image with dxCompiler and Java version 8 or 11 following the instructions in the dxCompiler Github repository
Install the Docker CLI
Use docker_run.sh script to invoke dxCompiler with the built Docker container image
More detailed setup instructions and sample WDL workflows can be found in the dxCompiler Github README.
Last updated