Ley Lab Data Analysis Workshop

A workshop covering the basic tools of bioinformatics

Install

Workshop material

Clone this repository:

git clone https://github.com/leylabmpi/data_analysis_workshops.git

Move to the repo directory:

cd data_analysis_workshops

Install miniconda & create conda environments

Note: this will install conda into the data_analysis_workshops directory. Thus, it will create a conda install (& conda environments) just for this workshop. If you already have a conda environment set up, This will override that other conda install, but just within the data_analysis_workshops directory.

Clone the conda_env_setup repository:

git clone https://github.com/leylabmpi/conda_env_setup.git

Move to the repo directory:

cd conda_env_setup

Follow the instructions listed in the README.md file to install conda & set up some conda environments:

cat README.md

Workshop materials

The workshop materials consist of Jupyter notebooks.

You can view the Jupyter notebooks on GitHub. See the notebooks directory

To start up Jupyter and run through the notebooks:

First, ssh onto the server using port fowarding:

ssh -L XXXX:localhost:XXXX USERNAME@rick.eb.local

Note: for the next command, change "XXXX" to a number between 7000 and 20000. Also, change "USERNAME" to your user name.

Move to the data_analysis_workshops directory.

Start up Jupyter:

screen jupyter notebook --no-browser --port XXXX

Note: make sure to change "XXXX" to the port you used for ssh-ing (described above). Note: the "screen" command was used so that Jupyter will still run even after you disconnect from the server.

Jupyter should start up. To get the command line back (detatch from screen), press: ctrl-a (then) d

Connect on your browser (preferably Chrome):

In your browser url box, type: "127.0.0.1:XXXX"

Note: "XXXX" is the port number you used previously

Material description

The tutorial notebooks are in ./notebooks/qiime_tutorial/.

All other notebooks will be used for a presentation, but feel free to run through those notebooks too!

Helpful code

See helpful_code.md