In this poster we showcase the recent developments in the GPU CHARMM code. The code is  written in CUDA and modern C++ and has a pubind11-based python interface for ease of use.The code supports the common methods for simulaiton in NVE,NVT and NPT ensembles, pre and prost-prcoessing, free energy calculations etc.  We especially focus on the adaptive sampling scheme in which we use normalizing flows to map the trajectory on the latent space and generate samples from the latent space to adaptively seed samples from the under sampled regions of interest. The python based interface allow easy integration of the code with moden deep learning framworks.

Recent development in machine learning based adaptive sampling methods for in GPU CHARMM

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.