Awesome

<p align="center"> <img src="docs/figs/ICoN.png"> </p>

Internal Coordinate Net (ICoN) for Sampling Conformational Ensembles of Highly Dynamic Proteins via Generative Deep Learning

ICoN is trained on ~10000 of fully atomistic and highly flexible conformations
It uses vector internal coordinate representation as input features- vBAT
It can train ~10K conformations in a few mins, and generate ~100K conformations in less than a min.

Dependancies

python 3.8 >
- pytorch - for Deep Learning
- MDAnalysis - for trajectory I/O
- pytraj - for analysis

Installation

Clone the repository: git clone https://github.com/chang-group/ICoN.git
Navigate to corresponding folder for further instruction