Improving the quality and interpretability of cryo-EM maps by local and non-local deep learning.

References:

• He J, Li T, Huang S-Y.* Improvement of cryo-EM maps by simultaneous local and non-local deep learning. Nature Communications, 2023;14:3217 Journal

• emready doc:doc

• Module Name: emready (see the modules page for more information)
• The emready (available on biowulf) command lines could be run as:

          EMReady.sh in_map.mrc out_map.mrc [Options]
  	

• emready testing input in $EMREADY_TEST_DATA

Allocate an interactive session and run the program.
Sample session (user input in bold):

[user@biowulf]$ sinteractive --gres=gpu:v100:1 --mem=20g
salloc.exe: Pending job allocation 46116226
salloc.exe: job 46116226 queued and waiting for resources
salloc.exe: job 46116226 has been allocated resources
salloc.exe: Granted job allocation 46116226
salloc.exe: Waiting for resource configuration
salloc.exe: Nodes cn3144 are ready for job

[user@cn3144]$ module load emready
[user@cn3144]$ mkdir /data/$USER/emready_test/
[user@cn3144]$ cd /data/$USER/emready_test/
[user@cn3144]$ cp $EMREADY_TEST_DATA/*.mrc .
[user@cn3144]$ EMReady.sh
EMReady.sh

USAGE: EMReady.sh in_map.mrc out_map.mrc [options]

Descriptions:
    in_map.mrc  : Input EM density map in MRC2014 format
    out_map.mrc : Filename of the output processed density map

    -g          : ID(s) of GPU devices to use.  e.g. '0' for GPU #0, and '2,3,6' for GPUs #2, #3, and #6
                  default: '0'

    -s          : The step of the sliding window for cutting the input map into overlapping boxes. Its value should be an integer within [12, 48]. A smaller stride means a larger number of overlapping boxes
                  default: '12'

    -b          : Number of boxes input into EMReady in one batch. Users can adjust 'batch_size' according to the VRAM of their GPU devices. Empirically, a GPU with 40 GB VRAM can afford a 'batch_size' of 30
                  default: '10'

    --use_cpu   : Run EMReady on CPU instead of GPU

[user@cn3144]$ EMReady.sh emd_22216.mrc emd_22216.out.mrc
# Running on 1 GPU(s)
# Loading pre-trained model...
# Voxel size of input map: [1.666 1.666 1.666]
# Using the EMReady model with grid size of 1.0 Angstrom
# Loading map data...
# Rescale voxel size from [1.666 1.666 1.666] to [1. 1. 1.]
# Map dimensions: [265 265 265]
# Split map into 12167 chunk(s)
# Get 12167 positive chunks
# Start prediction

[user@cn3144]$ exit
salloc.exe: Relinquishing job allocation 46116226
[user@biowulf]$

Create a batch input file (e.g. emready.sh). For example:

#!/bin/bash
set -e
module load emready
cp $EMREADY_TEST_DATA/*.mrc .
EMReady.sh emd_22216.mrc emd_22216.out.mrc -b 20 -g '0,1'

Submit this job using the Slurm sbatch command.

sbatch --gres=gpu:v100:2 --mem=40g emready.sh