🔴❗UNDER DEVELOPMENT❗🔴

Principal Component Analysis Automation (PCAauto)

PCAauto is a script developed to automate the PCA analysis of Molecular Dynamics.

Available methods are:

• PCA analysis
• Calculation of variance, cumulated varianc and explained variance (this last one only for 3 components for now)
• Calculation of PCA through simulation time
• Visualization o the movements of the components on model
• Also, there is the possibility to plot graphs (under development)

Instalation

Download the code and unzip it on the desirable directory. To prepare the environment use the following command:

conda env create -f environment.yml

Be aware to uncomment the sections on the environment.yml file depending on which OS you are using.

How to use

Activate the environment using:

conda activate pca-mdaanalysis

You may modify the main.py file depending on which calculation you are interested.

The program uses a shell script to automate the calculation of data in multiple folders. To use this feature go to the root directory and on the terminal use:

bash \calculatePCA_inBulk.sh

If you are interested only on running one folder, you may just add your files to the PCAauto directory and use:

python calculatePCA.py

Observations

This script was developed following the MDA Analysis documentation, specifically the MDA Analysis examples page.

Authorship

• Author: Brenda Ferrari (brendaferrari)