DAQ Score

DAQ Score is a computational tool using deep learning that can estimate the residue-wise local quality for protein models from cryo-Electron Microscopy (EM) maps.

Google Colab Version

Please use DAQ Colab.

Docker Container

To use the DAQ Docker image, you can either build the image on your own machine, or download the pre-built image.
To build your own image, please use the tutorial available at the DAQ Github repository.
The pre-built image can be downloaded from here. Please follow the rest of this guide to learn how to use it.

Docker Requirements and Installation Guides


(1) Nvidia driver available at


(2) Docker available at


(3) Nvidia container toolkit available at



(1) Nvidia driver available at


(2) Docker available at


(3) WSL 2: Instructions can be found at


(4) CUDA: Please follow the instructions at

https://docs.nvidia.com/cuda/wsl-user-guide/index.html, under Section 3. Cuda Support for WSL 2

(5) Open the Docker Desktop app and Enable WSL 2 Integration under Settings. Instructions available at


Usage guide

After installing the requirements, add the container to your Docker installation by issuing the following command:
sudo docker load --input /path/to/daq.tar

Now, you are ready to run the DAQ container. Place all your input files(map and structure file) in one directory; for example, if the directory is named “inputs”, it’ll contain the .pdb and .mrc files.
To run DAQ, you can issue the following command in the terminal:

sudo docker run --gpus=all -v /fullpath/inputs/:/DAQ-main/inputs daq --mode=0 --gpu=0 -F "2566_3J6B_9.mrc" -P "3J6B_9.pdb" --window 9 --stride 2

  • /fullpath/inputs/: Should be replaced with the full path to the inputs directory containing the input files.
  • --gpu: You can specify which GPU to use
  • -F: Map name
  • -P: Structure file
  • --window: half_window_size
  • --stride: stride size
After execution, your outputs will be saved in a new directory that will automatically be created inside the inputs directory.


© 2021 Genki Terashi, Xiao Wang, Sai Raghavendra Maddhuri Venkata Subramaniya, John J. G. Tesmer, and Daisuke Kihara, and Purdue University

GPL v3. (If you are interested in a different license, for example, for commercial use, please contact us.)