justin-cold / vina-gpu

A heterogeneous OpenCL implementation of AutoDock Vina

Vina-GPU

A heterogeneous OpenCL implementation of AutoDock Vina

A CUDA version of Vina-GPU is avaliable at here

Update: Vina-GPU+: a virtual-screening-accelerated version of Vina-GPU is released together with our Vina-GPU 2.0 toolkit

Compiling and Running

Note: at least one GPU card is required and make sure the version of GPU driver is up to date

Windows

Run on the executable file

1. For the first time to use Vina-GPU, please run Vina-GPU-K.exe with command ./Vina-GPU-K.exe --config=./input_file_example/2bm2_config.txt You are supposed to have the docking results 2bm2_out.pdbqt of our example complex and a Kernel2_Opt.bin file
2. Once you have the Kernel2_Opt.bin file, you can run Vina-GPU.exe without compiling the kernel files (thus to save more runtime)

When you run Vina-GPU.exe, please make sure Kernel2_Opt.bin file are in the same directory

For the usage and limitaiton of Vina-GPU, please check Usage and Limitation. A graphic user interface (GUI) is also provided for Windows users, please check GUI

Build from source file

Visual Studio 2019 is recommended for build Vina-GPU from source

1. install boost library (current version is 1.77.0)

2. install CUDA Toolkit (current version is v11.5) if you are using NVIDIA GPU cards

   Note: the OpenCL library can be found in CUDA installation path for NVIDIA or in the driver installation path for AMD

3. add ./lib ./OpenCL/inc $(YOUR_BOOST_LIBRARY_PATH)/boost $(YOUR_CUDA_TOOLKIT_LIBRARY_PATH)/CUDA/v11.5/include in the include directories

4. add $(YOUR_BOOST_LIBRARY_PATH)/stage/lib $(YOUR_CUDA_TOOLKIT_PATH)/CUDA/lib/Win32in the addtional library

5. add OpenCL.lib in the additional dependencies

6. add --config=./input_file_example/2bm2_config.txt in the command arguments

7. add WIN32 in the preprocessor definitions if necessary

8. if you want to compile the binary kernel file on the fly, add BUILD_KERNEL_FROM_SOURCE in the preprocessor definitions

9. build & run

Note: ensure the line ending are CLRF

Linux

Note: At least 8M stack size is needed. To change the stack size, use ulimit -s 8192.

1. install boost library (current version is 1.77.0)

2. install CUDA Toolkit (current version is 11.5) if you are using NVIDIA GPU cards

   note: OpenCL library can be usually in /usr/local/cuda (for NVIDIA GPU cards)

3. change the BOOST_LIB_PATH and OPENCL_LIB_PATH accordingly in Makefile

4. set GPU platform GPU_PLATFORM and OpenCL version OPENCL_VERSIONin Makefile. some options are given below:

   Note: -DOPENCL_3_0 is highly recommended in Linux. To check the OpenCL version on a given platform, use clinfo.

Macros	Options	Descriptions
GPU_PLATFORM	-DNVIDIA_PLATFORM / -DAMD_PLATFORM	NVIDIA / AMD GPU platform
OPENCL_VERSION	-DOPENCL_3_0 / -DOPENCL_1_2	OpenCL version 3.0 / 1.2

6. type make clean and make source to build Vina-GPU that compile the kernel files on the fly (this would take some time at the first use)

7. after a successful compiling, Vina-GPU can be seen in the directory

8. type ./Vina-GPU --config ./input_file_example/2bm2_config.txt to run Vina-GPU

9. once you successfully run Vina-GPU, its runtime can be further reduced by typing make clean and make to build it without compiling kernel files (but make sure the Kernel2_Opt.bin file is unchanged)

10. other compile options:

Options	Description
-g	debug
-DDISPLAY_ADDITION_INFO	print addition information

macOS

Note: The running Vina-GPU on macOS is not recommended and has not fully tested yet

1. install boost library (current version is 1.77.0)

modify Makefile as follows:

2. annotate OPENCL_LIB_PATH, OPENCL_INC_PATH and -L$(OPENCL_LIB_PATH)/lib64
3. add -framework OpenCL in LIB3
4. type make and run

Usage

Arguments	Description	Default value
--config	the config file (in .txt format) that contains all the following arguments for the convenience of use	no default
--receptor	the recrptor file (in .pdbqt format)	no default
--ligand	the ligand file (in .pdbqt fotmat)	no default
--thread	the scale of parallelism (docking lanes)	1000
--search_depth	the number of searching iterations in each docking lane	heuristically determined
--center_x/y/z	the center of searching box in the receptor	no default
--size_x/y/z	the volume of the searching box	no default

Limitation

Arguments	Description	Limitation
--thread	the scale of parallelism (docking lanes)	preferably less than 10000
--size_x/y/z	the volume of the searching box	less than 30/30/30

Graphic User Interface (GUI)

A graphic user interface (GUI) is provided for users on Windows OS

1. first make sure that Vina-GPU.exe can run on a terminal
2. put the Vina-GPU.exe and Kernel2_Opt.bin files in ./Vina-GPU/GUI/exec and overwrite the original files
3. run the Vina-GPU-GUI.exefile within ./Vina-GPU/GUI to start up the Vina-GPU GUI
4. select the input and out output files
5. set the box center, the box size, thread and search_depth
6. click the start button to run Vina-GPU

Citation

• Tang, S.; Chen, R.; Lin, M.; Lin, Q.; Zhu, Y.; Ding, J.; Hu, H.; Ling, M.; Wu, J. Accelerating AutoDock Vina with GPUs. Molecules 2022, 27, 3041. https://doi.org/10.3390/molecules27093041

• O. Trott, A. J. Olson, AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading, Journal of Computational Chemistry 31 (2010) 455-461.