Virtual drug screening with AutoDock Vina - Elastic High Performance Computing

Background information

Molecular docking is a key step in virtual drug screening. AutoDock Vina is an open-source molecular docking software that is fast, accurate, and ideal for building virtual screening workflows based on molecular docking. It is used with the MGLTools suite, which includes AutoDock Tools (ADT) and Python Molecular Viewer (PMV). ADT is used to generate input files for Vina, and PMV is used to view the results. For more information, see AutoDock Vina and MGLTools.

This topic demonstrates the simulation of interactions between small-molecule ligands and macromolecular receptors to predict their binding modes and affinities for drug screening. Commercial compound libraries, such as Specs, Enamine, and ChemDiv, provide a large number of ligands to simulate the interaction between ligands and a given receptor. Because the calculations for different ligands are independent of each other, they can be processed in parallel on a large scale. This method is also applicable to other scenarios in the biological and pharmaceutical fields that require large-batch and high-concurrency processing.

Prerequisites

An E-HPC cluster is created.

For more information, see Create a cluster using the wizard. Note the following parameters:

Parameter

Description

Hardware parameters

The deployment mode is Standard, which includes two management nodes, one compute node, and one logon node.

All nodes use the ecs.c7.large instance type, which provides 2 vCPUs, 4 GiB of memory, and a 2.7 GHz Ice Lake processor.

Software configuration

Select the CentOS 7.6 public image, select pbs as the scheduler, and enable VNC.

A cluster user is created. For more information, see Create a user.

You will use this user to log on to the cluster to perform operations such as compiling software and submitting jobs. The sample user for this topic is configured as follows:
- Username: vinatest
- Permission group: sudo
Install the software. For more information, see Install software.

The following software is required:
- vina, version 1.1.2
- AutoDockTools, version 1.5.7

Step 1: Connect to the cluster

Connect to the cluster using one of the following methods. This topic uses the username vinatest. After connecting, you will be in the /home/vinatest directory by default.

Connect by using the client
This method supports only clusters that use the PBS scheduler. Before you proceed, make sure that you have downloaded and installed the E-HPC client and configured the required environment. For more information, see Configure the client environment.
1. Start and log on to the E-HPC client.
2. In the left-side navigation pane, click Session Management.
3. In the upper-right corner of the Session Management page, click terminal to open the Terminal window.
Connect by using the console
1. Log on to the E-HPC console.
2. In the upper-left corner of the top menu bar, select a region.
3. In the left-side navigation pane, click Cluster.
4. On the Cluster page, find the target cluster and click Connect.
5. On the Connect page, enter the cluster username, password, and port, and click ssh connection.

Step 2: Submit the job

Download and extract the job files.
1. Download the job files.
  
  Note
  If Git is not installed on the cluster, run the sudo yum install -y git command to install it.
```
git clone https://best-practice:Abcd123456@codeup.aliyun.com/best-practice/bp/022.git
```
2. Extract the job files.
```
cd 022
tar xzvf vina-ehpcarrayjob.tar.gz
```
Modify the configuration file.
```
cd vina-ehpcarrayjob
vim conf.txt
```
In the conf.txt file, delete the following two lines:
```
receptor = 1fkn_rgd.pdbqt
ligand = $file
```

Create a job script named vina.sh.

cd /home/vinatest
vim vina.sh

The script contains the following content:

#!/bin/bash
#PBS -N vina_job
#PBS -l nodes=1:ppn=2
#PBS -l walltime=00:10:00
#PBS -j oe
#PBS -v receptor="022/vina-ehpcarrayjob/1fkn_rgd.pdbqt",ligand="022/vina-ehpcarrayjob/test/ligand_1.pdbqt",config="022/vina-ehpcarrayjob/conf.txt"
cd $PBS_O_WORKDIR
ppn="$NCPUS"
cd $PBS_O_WORKDIR
vina --receptor $receptor --config $config  --ligand $ligand --out out.pdbqt --cpu $ppn

Submit the job.
```
qsub vina.sh
```
The expected output is the job ID 0.scheduler.
```
0.scheduler
```

Step 3: View the results

Check the job status.
```
qstat -x 0.scheduler
```
In the output, if the value of S is R, the job is running. If the value of S is F, the job is complete.
```
Job id            Name             User              Time Use S Queue
----------------  ---------------- ----------------  -------- - -----
0.scheduler       vina_job         vinatest          00:01:56 F workq 
```
Note
When the job finishes, it generates the result file out.pdbqt in the submission directory. For this example, the file is located at /home/vinatest/out.pdbqt.
Use VNC to visualize the job results.
1. Enable VNC.
  Note
  Make sure that the ports required by VNC are enabled for the security group to which the cluster belongs. When you use the console, the system automatically enables port 12016. When you use the client, you must enable ports manually. The first VNC window uses port 12017. If multiple users need to use VNC, port numbers increment sequentially.
  - Use the client
    1. In the left-side navigation pane, click Session Management.
    2. In the upper-right corner of the Session Management page, click VNC to open VNC Viewer.
  - Use the console
    1. In the left-side navigation pane of the E-HPC console, click Cluster.
    2. On the Cluster page, find the target cluster and click More > VNC.
    3. Use VNC to remotely connect to the visualization service. For more information, see Connect to the visualization service.
2. In the VNC window, choose Application > System Tools > Terminal.
3. In the Terminal, run the following command to open AutoDock Tools and load the job results.
```
/usr/local/bin/adt /home/vinatest/022/vina-ehpcarrayjob/1fkn_rgd.pdbqt /home/vinatest/out.pdbqt
```
  After the model loads in the AutoDock Tools window, the results are displayed.
  
  The molecule list on the left side of the AutoDock Tools window displays 10 entries: 1fkn_rgd and out_model1 to out_model9. The main viewport displays a 3D visualization of all docked conformations as a superimposed stick model on a black background.