Skip to content
FireCloud Service Selector (FISS) -- Python bindings and CLI for FireCloud execution engine
Python Shell Makefile Other
Branch: master
Clone or download
dheiman v0.16.25 (#129)
Multiple enhancements to mop:
* Only delete unreferenced files generated by workflows. (#127 by @Marianie-Simeon)
* Ignore more files generated by the execution engine (e.g. rc, script, stdout, stderr). (closes #128)
* Report file sizes in summaries and listings. (closes #118)
* Allow user to include/exclude files via glob. (closes #116)
Latest commit 3870bee Aug 2, 2019
Type Name Latest commit message Commit time
Failed to load latest commit information.
firecloud v0.16.25 (#129) Aug 2, 2019
plugins Bugfixes for api & attr_fill_null, new flow_submit fiss cmd Oct 20, 2016
util initial version Jul 6, 2017
.dockerignore FISS v0.16.13: Jun 1, 2018
Dockerfile FISS v0.16.13: Jun 1, 2018
LICENSE FISS needs a license. May 9, 2018
Makefile v0.16.22 (#122) Apr 3, 2019
README Release Candidate v0.16.12 May 30, 2018
changelog.txt v0.16.25 (#129) Aug 2, 2019
requirements.txt initial conversion to using google.auth rather than oauth2client Aug 25, 2017 pypi now requires explicit content type for long_description Jun 8, 2019


1. Description

FISS -- (Fi)reCloud (S)ervice (S)elector

FISS is a programmatic interface to FireCloud (FC), providing a set of low- and
high-level Python bindings to the FireCloud API, as well as UNIX bindings for
command line usage.  By wrapping the FireCloud RESTful API in this manner, our
hope is to provide an interface that resonates more closely with the majority of
expected FC users--supporting interaction with FC in memes familiar to them, as
biomedical researchers & informaticians rather than database or web programmers

Like legacy FISS, the (Fi)rehose (S)ervice (S)selector that was created for
internal use at the Broad Institute, FISSFC aims to be:

    fast               compared to navigating screens & menus in the GUI,
                       importing/exporting files manually, etc
    programmatic       can be used in analysis scripts, UNIX cli, cron jobs etc
    iterable           easily repeat commands across workspaces, entity sets,
                       workflows, and even project domains
    simple to use      don't have to learn FC internal abstractions, or code
                       directly to the RESTful interface (e.g. in curl)
    understandable     clear & consistent semantics, for FC novice or expert
    extensible         users can customize how it works, without permission
                       (or needing development assistance) from the author!
    portable           Python 2 and 3 implementation, UNIX shell
    idiomatic          allows FireCloud to be combined with the power of UNIX
    self-documenting   (well, almost!) in addition to the --help flag, usage
                       statements are provided for most methods when they're
                       called with no args; the -F flag can display the body
                       of any method, and -l flag lists all available methods

2. Contents
1. Description
2. Contents
3. Requirements
4. Unix command line
5. Low-level interface
6. High-level interface
7. User configuration

3. Requirements

Install this package via pip:
% pip install firecloud

Or download the source and install with
% python install

FISS depends on several additional Python software packages, as listed in; typically the entire installation process (including loading of
packages) takes only a few minutes.

**Note: FireCloud encryts all web traffic with TLSv1.2. Some versions of
python will not function correctly, and must be recompiled with openssl
v1.02 or later. If recompiling is not an option, there is also a
dockerized version of the tool available on docker hub

% docker pull broadgdac/fiss

4. UNIX command line interface

After installation your the 'fissfc' executable should be findable in your
$PATH.  Type 'fissfc --help' for more details on how to use this utility
in a UNIX setting.

Note that if you are running on a Google VM, executing 'pip install' as an
unprivileged user (i.e. not sudo) will result in the package being installed
to a directory named $HOME/.local, so you will need to ensure $HOME/.local/bin
is in your $PATH (e.g. by setting it in your Bash login script) so that
the fissfc binary can be found.

5. Low-Level API

The low-level API provides a direct one-to-one mapping between the
RESTful API of FireCloud and methods in the Python interpreter. It
also provides conveniences like automatic header creation, paging,
and debug settings (e.g. verbosity) across the entire API.  To access
the low-level API simply import into a Python interpreter:  for example

    import firecloud.api as fapi

After importing, you can see what low level methods are available via


and for detailed documentation simply use the Python help system:


Finally, note that interactive documentation for the FireCloud RESTful API
(sometimes referred to as the Orchestration Layer) can be found online at

6. High-Level Classes

To access the high-level interface import the 'fiss' module, e.g.

    from firecloud import fiss

and use the dir() and help() Python as shown above.

7. User configuration

The behavior of FISS can be tailored by writing a configuration file,
e.g. to give default values for the project and workspace parameters
that are common to many operations with FireCloud.  This would make it
easier to use the UNIX command line interface by allowing the --project
and --workspace flags to be omitted.

By default FISS will look for a user-specific configuration file at


For example, developers within the Broad Genome Data Analysis Center (GDAC) 
use a configuration file like


To inspect your runtime configuration, simply issue the config command:

	linux%    fissfc  config
	debug   		False
	entity_type     sample_set
You can’t perform that action at this time.