WindowsInstallation.md

Local Installation on Windows

If you have any problems with this installation, please file an issue and describe any problems, so we can improve the instructions.

Install the lastest version of Python using Anaconda

Download the Python 3.7 Anaconda installer and run the Anaconda installer.

Then go to the Start Menu and open Anaconda Prompt

For easy access, pin the Anaconda Prompt to the task bar.

Install Git

The Git version control system is used to download repositories from Github.

Skip this step if you have git installed.

Download Git and run the git installer (choose all default options).

Clone this repository

git clone https://github.com/sbl-sdsc/mmtf-proteomics.git

Set the following environment variables

setx SPARK_CONF_DIR <your path>\mmtf-proteomics\conf

setx HADOOP_HOME <your path>\mmtf-proteomics\conf

Important: Close the Anaconda Prompt and reopen it to set the environment variables.

Create a conda environment

cd mmtf-proteomics

conda env create -f binder/environment.yml

Activate the conda environment

conda activate mmtf-proteomics

Launch Jupyter Notebook

jupyter notebook

After you are finished, deactivate the conda environment

conda deactivate

Anytime you want to use the environment, activate it again and start Jupyter Notebook

To permanently remove the benchmark environment

conda remove -n mmtf-proteomics --all

Setting Spark Configurations

When running PySpark on many cores (e.g., > 8), the memory for the Spark Driver and Workers may need to be increased. To change memory setting, go to the mmtf-proteomics\conf folder and edit the file spark-env.cmd. By default, this file has the following settings:

SPARK_DRIVER_MEMORY=4G
SPARK_WORKER_MEMORY=4G

When running this repo on 24 core machine, you may need to increase the memory settings to 20G.