autodock-GPU how to set up in windows system.

[necat61]

In the Windows 11 Pro system, I downloaded the AutoDock-GPU source code from GitHub of the latest release, and I used autodock-gpu.sln to build the autodock-gpu.exe file using Visual Studio 2022. So, I can use autodock-gpu with cmd; if I write autodock-gpu, it's showing how to use it. But I can't be sure if I've set up correctly or not. In Visual Studio, I set the build configuration to Release and the platform to x64. and I opened the autodock-gpu.sln file with Visual Studio. Right-click on the autodock_gpu in the Solution Explorer and select "Properties." Under "Configuration Properties" > "VC++ Directories," set the Include Directories to the path of my CUDA include directory (e.g., C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6\include). Set the Library Directories to the path of your CUDA library directory (e.g., C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6\lib\x64). Right-click on the autodock_gpu project in the Solution Explorer and select "Build.". I prepared the autodock-gpu.exe file this way, and I couldn't be sure because there are 506 warnings as output while building the executable file in Visual Studio. Please, I'd like you to provide me information about whether I've done correctly set up or not.

[necat61]

‘=’: ‘uint64_t’ to ‘uint32_t’ conversion; data may be lost ‘/openmp’ is overridden by ‘/openmp:llvm’ ‘argument’: ‘time_t’ to ‘uint32_t’ conversion; data may be lost ‘initialising’: truncated from ‘double’ to ‘float’

‘strncpy’: This function or variable may be unsafe. Consider using strncpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.

‘=’: ‘size_t’ to ‘int’ conversion; data may be lost

‘strdup’: The POSIX name for this item is deprecated. Instead, use the ISO C and

C++ conformant name: _strdup. See online help for details.

‘=’: ‘size_t’ to ‘int’ conversion; data may be lost

‘=’: ‘size_t’ to ‘unsigned int’ conversion; data may be lost

‘strdup’: The POSIX name for this item is deprecated. Instead, use the ISO C and C++ conformant name: _strdup. See online help for details.

‘argument’: ‘size_t’ to ‘_Ty’ conversion; data may be lost

‘argument’: ‘time_t’ to ‘uint32_t’ conversion; data may be lost

‘=’: ‘_Ty’ to ‘unsigned int’ conversion; data may be lost

‘=’: ‘const double’ to ‘float’ conversion; data may be lost

‘/LTCG’ is ignored due to ‘/INCREMENTAL’ specification uninitialised local variable ‘x_frac’ used uninitialised local variable ‘x_high’ used uninitialised local variable ‘y_frac’ used uninitialised local variable ‘y_high’ used uninitialised local variable ‘z_frac’ used uninitialised local variable ‘z_high’ used fopen': This function or variable may be unsafe. Consider using fopen_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.

Mainly warnings that I encountered when I built autodock-gpu.exe file. (as output of visual studio 2022) Help me to understand whether ı set up true or not.

[atillack]

Here is my VS2022 output when compiling using Build->Build AutoDock-GPU:

1>AutoDock-GPU.vcxproj -> C:\Users\Andreas Tillack\source\repos\AutoDock-GPU-Release\bin\AutoDock-GPU.exe
1>Done building project "AutoDock-GPU.vcxproj".
========== Build: 1 succeeded, 0 failed, 0 up-to-date, 0 skipped ==========
========== Build completed at 13:23 and took 03.864 seconds ==========

The Windows build currently uses OpenCL so you don't need to change and add Cuda directories (although it shouldn't hurt but it'll still compile for OpenCL currently).

[necat61]

I couldn't understand what to do because I'm so new to code things. Please tell me how to set up using Visual Studio 2022. what could I add to library and include while I'm in the sln. file. I opened sln file with Visual Studio 2022 and what is the last???

[atillack]

The compilation should work as-is with no need to add anything. Opening AutoDock-GPU.sln and then using Build->Build AutoDock-GPU is all that is needed to produce AutoDock-GPU.exe in the bin folder. So if you ended up with bin\AutoDock-GPU.exe you should be good to go.

[necat61]

I can build succesfully but it occurs 506 warning when I built with this way. If you say you don't need to pay attention to this warnings (as you can see above). The warnings like;

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\inc\miscellaneous.h(214,8): warning C4244: ‘=’: ‘uint64_t’ to ‘uint32_t’ conversion; data loss may occur

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\inc\getparameters.h(94,62): warning C4305: ‘initialising’: truncated from ‘double’ to ‘float’

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\src\calcenergy.cpp(208,45): warning C4244: '=': 'double' - 'float' dönüşümü; veri kaybı olabilir

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\src\getparameters.cpp(213,16): warning C4996: ‘strcpy’: This function or variable may not be safe. Consider using strcpy_s instead. Use _CRT_SECURE_NO_WARNINGS to disable deprecation. See the online help for details.

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\src\getparameters.cpp(214,37): warning C4996: 'strdup': The POSIX name for this item is deprecated. Instead, use the ISO C and C++ conformant name: _strdup. See online help for details.

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\src\getparameters.cpp(668,12): warning C4996: 'sscanf': This function or variable may be unsafe. Consider using sscanf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.

1>C:\Users\Hp\Desktop\GPU\AutoDock-GPU-1.6\AutoDock-GPU-1.6\host\src\processligand.cpp(1713,41): warning C4244: '=': 'const double' - 'float' transformation; data loss may occur

does it occur any warnings while you are building ? or you've just send the out of visual studio last part. if you say you don't need to pay attention to these warnings, I can ignore them and go on.

[atillack]

Yes, I do see compile warnings too and for the moment think they are not show stoppers as it's mostly implicit type conversions and the usual suspects of Posix functions. That said, I will eventually lower the number of them (even if it means selectively disabling some of them, i.e. for implicit type conversions).

[Jeffiq]

I got the File I extracted into Folder opened AutoDock-GPU.sln, Visual Studio Opened and I ran selected Build From Ribbon of Visual Studio a drop down Menu Appeared and I selected Build AutoDock-GPU Ctrl + B and it ran Output I had
1>Done building project "AutoDock-GPU.vcxproj" -- FAILED. ========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ========== ========== Build completed at 00:33 and took 16,546 seconds ==========

and other long Stuff above that Error List I had 28 Errors and 502 Warnings

[atillack]

@Jeffiq Thank you for reporting. This is caused by the "Debug" version accidentally not having C++ 17 set. Easy fix which I'll update soon.

The solution is to compile the "Release" version which I would recommend in either case as the debug version will slow performance and has a lot of output for, well, debugging :-)

[necat61]

I have to ask how to handle the protein preparation when it contains HEM or HEME in its active site (it have to be hold since the effectivness of potential drug molecule addicted to this) The python script provided by autodocktools assign zero charge to FE atom. Could you break it down how to sort out this problem ? I may use Autodock-GPU, Autodock 4.2 or Vina ( I downloaded just now).