Note: currently driverMAPS can only be run on Linux machines.

Step 1: Installing Miniconda 3

To install Miniconda on your linux machine, please follow the official guidelines here. As of 2019 April, you can do it like this: assuming that you have a 64-bit system, on Linux, download and install Miniconda 3 with


For a 32-bit system, URLs and file names are analogous but without the _64. When you are asked the question:

Do you wish the installer to prepend the Miniconda3 install location to PATH …? [yes|no] answer with yes.

Step 2: Preparing a working directory and download driverMAPS

First, create a new directory by mkdir path/to/workdir at a reasonable place then download driverMAPS and uncompress it to your working directory

tar -xf driverMAPS-v1.0.x.tar.gz -C path/to/workdir
cd path/to/workdir

(Note: replace driverMAPS-v1.0.x.tar.gz with the exact file name you downloaded)

After uncompression, the folder takes around 7G, so please make sure you have enough space before uncompress the software.

Step 3: Creating an environment with the required software

The environment.yaml file can be used to install all required software into an isolated Conda environment with the name maps. This step takes about 5min.

cd envs/
conda env create --file environment.yaml

(2019.10.05, I received reports from some users that they could not installed with the provided environment.yaml file and need to change a line - requests=2.14.2=py36_0 to - requests=2.13.*=py36_0 to successfully install. I am investigating this issue.)

Then install package “hmmfb” (which is not on conda)

conda activate maps
R -e 'library(devtools);options(unzip = "internal");install_github("simingz/hmm_fb")'
conda deactivate

If you see errors occur when installing “hmmfb” using the above command, you can also try install from the source code like this:

conda activate maps
# download and unzip source code from or use the following command:
git clone
# Build and install
R CMD build --no-build-vignettes hmm_fb/
R CMD INSTALL hmmfb_1.0-4.tar.gz
conda deactivate

Note: if your conda version is before 4.4.0 (released December 2017), you should use source activate maps instead of conda activate maps, and source deactivate instead of conda deactivate in the above commands.

This R Markdown site was created with workflowr