UPDATED TUTORIAL HERE. This is to install Alphafold v.2.1.0, and will no longer be updated. No idea how Alphafold is set up on your HPC? Want to run your own version of Alphafold? Well, you've come to the right place! This tutorial will show you how to install your own Alphafold, manage it using a Conda environment, and run a prediction of the protein, Bax. The following steps will be performed solely on OSCER (~3.5 Gb of disk space). Setting up your Conda Environment1. First we'll need to log in to OSCER. ssh <username>@schooner.oscer.ou.edu 2. Now that you're in your home directory, let's make a folder called "Programs," and change directories to the new folder. We will install Miniconda and Alphafold here. mkdir Programs cd Programs/ 3. To install minconda, follow the hyperlink and you should see the Miniconda3 Linux 64-bit version. We'll use the command "wget" to download the file directly to our Programs folder on the OSCER account. (Alternatively, if you downloaded this to your local computer, you can also copy the file to OSCER with "scp" or a file manager). Once the download finishes, we will start the installation with the command "bash." wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh bash Miniconda3-latest-Linux-x86_64.sh 4. Follow the prompts on the installer screen. Close and re-open the terminal, and test if your conda is working by typing the following command. This will list any packages that were installed by default. conda list 5. We will now make a virtual environment, to install any packages needed by Alphafold. More info on virtual environments can be found here. We'll name the new environment, "alphafold," and use Python 3.8. Once an environment has been made, we need to activate the environment. The base environment should now show the alphafold environment. conda create -n alphafold python=3.8 conda activate alphafold Setting up Tensorflow and GPU |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |