GrandPrix is a package for non-linear probabilistic dimension reduction algorithm in python, using TensorFlow and GPFlow. GrandPrix uses sparse variational approximation to project data to lower dimensional spaces. The model is described in the paper

"GrandPrix: Scaling up the Bayesian GPLVM for single-cell data.", Sumon Ahmed, Magnus Rattray and Alexis Boukouvalas, bioRxiv, 2017.

To replicate the results in the paper please use the betaVersion branch. The master branch works with the latest version of GPflow.

N.B. The package contains several large data files which are needed to run the example notebooks. Please be sure that your system has Git Large File Storage (Git LFS) installed to download these large data files.

If you have any problems with installation see the script at the bottom of the page for a detailed setup guide from a new python environment.

• Install tensorflow

pip install tensorflow

• Install GPflow

git clone https://github.com/GPflow/GPflow.git
cd GPflow    
pip install .
cd

See GPFlow page for more detailed instructions.

• Install GrandPrix package

git clone https://github.com/ManchesterBioinference/GrandPrix
cd GrandPrix
python setup.py install
cd

To run the notebooks

cd GrandPrix/notebooks
jupyter notebook

When running GrandPrix in a cluster it may be useful to constrain the number of cores used. To do this insert this code at the beginning of your script.

from gpflow import settings
settings.session.intra_op_parallelism_threads = NUMCORES
settings.session.inter_op_parallelism_threads = NUMCORES

• Create a new environment

conda create -n newEnv python=3.5

• Activate the new environment

source activate newEnv

• Create a new directory

mkdir newInstall
cd newInstall

• Follow the regular installation process described above