Linux Software Installation – Exercises 1

Part 1. Setup Conda

1.1 Login to the BioHPC machine and download the installer

Login (ssh) to the machine that you are assigned for this workshop (assigned machines: https://biohpc.cornell.edu/ww/machines.aspx?i=128). Prepare the working directory, and download the latest version of Miniconda/python3 installer into the working directory.

xxxxxxxxxx
mkdir /workdir/$USER
cd /workdir/$USER
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh 

1.2 Run the installer

xxxxxxxxxx
sh Miniconda3-latest-Linux-x86_64.sh

• The “chmod u+x” command makes the file executable.

• During the installation, you will be asked for multiple questions:

  1. “Do you accept the license terms?”: Type "yes" and press “ENTER” key;

  2. “Miniconda3 will now be installed into this location … ”: The default answer is "/home/$USER/miniconda3". You can either press "Enter" key to accept default or enter a different installation location. If your lab has a hosted BioHPC server, we strongly encourage you to enter "/workdir/xxxxx/miniconda3" (xxxxx is your userID) to install in local directory. This would make software installation/loading more than 10x faster.

  3. “Do you wish to update your shell profile to automatically initialize conda?” : Press ENTER to accept the default “no”;

1.3 Miniconda is ready.

Part 2. Install and run a simple software in Conda

2.1 Activate Conda

If you installed Conda in your home directory

x
source ~/miniconda3/bin/activate

If you installed Conda in /workdir/$USER directory

x
source /workdir/$USER/miniconda3/bin/activate

2.2 Install and run bwa in Conda

x
conda install bwa

bwa

• While you are at this step, you might want to use “which” command to check which copy of “bwa” you are running.

xxxxxxxxxx
which bwa

Part 3. Work with Conda environment

3.1 Create a virtual environment and give it a name “pysam”, install pysam the virtual environment

xxxxxxxxxx
conda create -c bioconda -n pysam pysam

• The command “conda create -n aligners” would create an environment called “pysam” (you can use any names), and install the pysam package into it.

3.2 Start the pysam environment, and run python

xxxxxxxxxx
conda activate pysam
which python
python -V
which pip

• “which python” “which pip”would tell you which python and pip you are using .

   

3.3 Install other python modules in this environment

xxxxxxxxxx
 pip install numpy

• The “pip” command would install numpy in the environment.

3.4 End the environment

xxxxxxxxxx
conda deactivate

3.5. Now you installed pysam and numpy in miniconda3 in the home directory. Next time you need to use it in a new session, run these commands

xxxxxxxxxx
source $HOME/miniconda3/bin/activate
conda activate pysam
# after the work is done
conda deactivate

Part 4. Create a Python3.12 environment in Miniconda3

xxxxxxxxxx
conda create -n myNewPipeline python=3.12
conda activate myNewPipeline
which python
python -V    ##use  capital V
conda deactivate 

Note in the first command there is no package name. This step would create an empty python 3.12 environment, within which you can use pip to install other python modules.