Requirements
If you are not in sudoers, then you can still install AMO without root access. Firstly, you are required to install the dependencies step by step.
Unix-like system
sudo apt-get update
sudo apt-get install -y --no-install-recommends \
build-essential \
libgflags-dev \
libgoogle-glog-dev \
libomp-dev \
libeigen3-dev \
libbz2-dev \
libhdf5-dev
Mac OSX
We highly recommend using the Homebrew package manager to install the dependencies.
brew install -vd cmake gflags glog boost
brew install zlib bzip2Step-by-step
GCC
Sometimes, on an old UNIX-like distribution, a new version of gcc is required (i.e. utilize the features from C++11 standard) before installing other dependencies. Don't worry, here is how you can compile your own gcc on the distribution with/without root access.
Make
If the version of gcc is not too old on your distribution, you can just simply install the make as follows,
If you compile your gcc own , then you are recommanded (as shown above in step 5, just uncomment the lines) to run as follows,
Cmake
Or install cmake by following lines if you compile your own gcc
Boost library
On Unix-like system, it is easy to install Boost library with sudo, type the following command in the terminal to install Boost library.
If you would like to install the Boost library without root, you can follow this instruction to install it.
Gflags
The gflags package contains a C++ library that implements command line flags processing.
The latest version of gflags is not validated for the AMO software.
Glog
The glog library contains a C++ implementation of the Google logging module.
The latest version of glog is not validated for the AMO software.
HDF5
AMO is only tested on HDF5 Version 1.8.14
PyTorch
More details can be found here. In AMO, we call modules in PyTorch to build deep learning models that are leveraged to analyze biological data.
Plotting
If one would like to run amo with figures, plotting packages are required to install. Gnuplot and R are two packages to generate figures.
Then, install library circlize as follows,
Install Gnuplot in the terminal as follows,
Last updated