• You can absolutely use the version of python installed on your OS if you are able to install packages. Sometimes, python is integrated with an OS, and therefore requires sudo to install packages because the package folder is in the root directory.

• If you have "pip" and "virtualenv" in your system's python distribution, then it is easiest to simply continue with that distribution and not worry about downloading another.

• Alternatively, if you are able to install packages to a location that does not require sudo permission, this is also fine. Again, you just need "pip" and "virtualenv"

• If you are not able to install pacakges due to permissions, then install the following.

  • Go to http://continuum.io/downloads#all and download the version of anaconda python 2.7.8 for your operating system.
  • Set up python
  • Add "export PATH='/path/to/anaconda/bin:$PATH" to your .bash_profile. You may need to similarly tell the system about a file called LD_LIBRARY.....
  • Once "pip" and "virtualenv" are installed, follow the instructions below to install OMOP and Harvest.

Harvest is a toolkit for building web applications that facilitates integrating, discovering, and reporting on data. It is developed and maintained by the Center for Biomedical Informatics (CBMi). It is nationally funded, designed for biomedical data first, open source and available on GitHub, and it comes with an HTML5 web client.

For best performance, please use Chrome, Firefox or Safari.

These instructions are intended for use with a PostgreSQL database.

virtualenv omop_harvest_env

cd omop_harvest_env
git clone https://github.com/rv816/omop_harvest.git

source bin/activate
cd omop_harvest
pip install -r requirements_grdr.txt

• Contact chopteam@ for further db instructions.

In the root directory of the repository ("omop_harvest"), you will find ".project_config.json" file (use ls -a to reveal the file. For omop projects, this file is where you should change the user/password for the database once you are set up.

• Insert a unique secret key (can be generated at http://www.miniwebtool.com/django-secret-key-generator/)

Again, if you are using your own data or a non-PGSQL database, it is especially important that you check the Django models defined in omop_harvest/models.py. Specifically, check to be sure the db_table setting references a real table for each model and that each field references a real column on that table. Notice, however, that ForeignKey fields will be named with the _id at the end of the column name truncated.

We run our apps using an nginx server that passes requests to uWSGI processes managed by supervisord and include server settings to that effect in the server directory. You should do whatever is most comfortable for you. If you don't want to bother with a production-type server environment right now, just do python bin/manage.py runserver 5678 and then open your browser and navigate to http://localhost:5678.