Installation with Conda

Prerequisites

• Conda (https://conda.io)

Preparation of conda evironment

Conda provides isolated environments in which we can install software packages to run separately from other software installed on the system. ssHMM depends on several other software packages which we will install into a brand new conda environment in the next steps.

1. Prepare the .condarc configuration file in your home directory (create if it does not exist)

  channels:
- bioconda
- defaults
- conda-forge

This step makes sure that the required dependencies are downloaded and installed from the correct source repositories (i.e. conda channels). Conflicts have been reported between software packages from different conda channels.

2. Create a new conda environment sshmm_env with all required software packages

conda create --name sshmm_env python=2.7 pip numpy forgi=1.1 graphviz pygraphviz rnashapes=2.1.6 rnastructure ghmm

3. Find out in which directory the new environment was created in

conda env list | grep sshmm_env

4. Make sure that an important environment variable is set for the RNAstructure prediction tool every time the environment is activated. Replace <ENVDIR> with the directory you found out in the previous step.

cd <ENVDIR>
mkdir -p ./etc/conda/activate.d
mkdir -p ./etc/conda/deactivate.d
echo -e '#!/bin/sh\n\nexport DATAPATH=<ENVDIR>/share/rnastructure/data_tables/' > ./etc/conda/activate.d/env_vars.sh
echo -e '#!/bin/sh\n\nexport DATAPATH=' > ./etc/conda/deactivate.d/env_vars.sh

Installation of ssHMM

5. Activate the new conda environment

source activate sshmm_env

6. Install ssHMM

pip install sshmm