🐲
DRGN-AI Guide
  • Introduction
  • Getting started
  • Running a job
  • Full Documentation
Powered by GitBook
On this page
  • Inputs/Outputs
  • Method Overview
  • Tutorial Outline
  • Citing this Work

Introduction

Welcome to the documentation for DRGN-AI!

NextGetting started

Last updated 11 months ago

DRGN-AI is an open-source software package for ab initio heterogeneous reconstruction in cryo-EM. This user guide is a work-in-progress. Please check back regularly for updates!

Inputs/Outputs

DRGN-AI takes a collection of picked particles with estimated CTF parameters as input and outputs an approximation of the conformational space of the biomolecular complex. The output takes the form of a set of low-dimensional latents (one for each picked particle) and a neural network that associates latents to 3D density maps.

Method Overview

DRGN-AI needs to jointly infer the pose of each particle in the dataset while reconstructing the structural variability of the molecule. To do so, it fuses the flexibility of implicit neural representations with a fast and scalable pose estimation strategy. Poses are optimized using a hierarchical search strategy and are later refined by stochastic gradient descent. Conformations are independently optimized in an autodecoding fashion. The whole model (neural field + conformations + poses) is optimized by gradient descent in order to minimize the difference between reconstructed images and real observations.

Tutorial Outline

The user guide is split into 3 parts:

  1. Installation: details on how to install DRGN-AI.

  2. Running a job: tutorial explaining how to run DRGN-AI, monitor your job and analyze the outputs.

  3. Full API documentation: details on all the DRGN-AI parameters.

Citing this Work

@article{drgnai,
  title    = "Revealing biomolecular structure and motion with neural ab initio
              {cryo-EM} reconstruction",
  author   = "Levy, Axel and Grzadkowski, Michal and Poitevin, Frederic and
              Vallese, Francesca and Clarke, Oliver B and Wetzstein, Gordon and
              Zhong, Ellen D",
  journal  = "bioRxiv",
  pages    = "2024.05.30.596729",
  month    =  jun,
  year     =  2024,
  language = "en"
}