Installation¶

Using conda (recommended)¶

The recommended way to install IM-Fusion is using conda, as this allows us to install IM-Fusion together with its external dependencies (tophat2, bowtie and subread) into an isolated environment using a single command:

conda create -n imfusion -c jrderuiter -c bioconda -c r im-fusion

Note that this uses the bioconda channel to install the external dependencies (tophat2, bowtie and subread) and the r channel to install R and rpy2.

Alternatively, IM-Fusion can be installed in an existing environent using:

conda install -c jrderuiter -c bioconda -c r im-fusion

Using pip¶

IM-Fusion can also be installed using pip:

pip install git+https://github.com/jrderuiter/im-fusion.git

Note that in this case, rpy2 is not installed by default as it can be difficult to install and is only required for the single-sample differential expression test. To use the test, ensure a working version of rpy2 is installed or install the required rpy2 dependencies using:

pip install git+https://github.com/jrderuiter/im-fusion.git[de_single]

To also install the development dependencies of IM-Fusion use:

pip install git+https://github.com/jrderuiter/im-fusion.git[dev]

Dependencies¶

Python dependencies¶

IM-Fusion is compatible with both Python 2.7 and Python 3.4+. However, identifying insertions only works in Python 3.4+ if a Python 3 compatible version of Tophat2 is installed. Such a version can be obtained from bioconda or by applying the 2to3 conversion to Tophats scripts.

IM-Fusion was tested using the following versions for its dependencies:

future – 0.15.2
toolz – 0.7.4
pandas – 0.18
numexpr – 2.5
scipy – 0.17
seaborn – 0.7
pysam – 0.8.4
intervaltree – 2.1.0

The following additional dependencies are required for running tests and building the docs:

pytest
pytest-mock
pytest-datafiles
pytest-helpers-namespace
sphinx

External dependencies¶

IM-Fusion has three external dependencies:

Tophat2 – for the alignment and detection of gene fusions
Bowtie1 – used by Tophat2 for the alignment
featureCounts – for generating expression counts

These tools must be available in the current PATH for the insertion detection to work properly. For help with their installation, see below.

Tophat2¶

Tophat2 is a RNA-seq read aligner that also includes Tophat-Fusion, an aligner that can identify gene fusions from single- and paired-end RNA-sequencing data. A binary version of Tophat2 can be downloaded from the Tophat website. After unpacking, make sure that the binary Tophat2 is available in PATH.

Alternatively, Tophat2 can also be installed using bioconda, or using homebrew or linuxbrew. For homebrew/linuxbrew, first install homebrew or linuxbrew and then install Tophat2 using:

brew update
brew install tophat

Bowtie¶

Bowtie can be installed in the same fashion as Tophat, by downloading a binary version from the bowtie website and ensuring that bowtie is available in PATH. Bowtie can also be installed using bioconda or using homebrew/linuxbrew:

brew update
brew install bowtie

featureCounts¶

featureCounts is an efficient tool for generating expression counts from RNA-seq alignments. It is part of the subread package, which can be downloaded from sourceforge. After downloading and unpacking the correct binary version for your machine, make sure the binary featureCounts is available in PATH.

Subread can also be installed using bioconda, or by using homebrew or linuxbrew with a custom tap:

brew tap jrderuiter/tap
brew update
brew install subread