Unable to dock

[pissurlenkar]

Input settings

/opt/unidock --receptor 5ikr-cox2.pdbqt \ --gpu_batch PV-002891686951.pdbqt PV-003124505859.pdbqt PV-003478069427.pdbqt \ --search_mode balance \ --scoring vina \ --center_x -32.29 \ --center_y 43.44 \ --center_z -5.64 \ --size_x 20 \ --size_y 20 \ --size_z 20 \ --num_modes 1 \ --dir results/ \

Verbose:

Rigid receptor: 5ikr-cox2.pdbqt Grid center: X -32.29 Y 43.44 Z -5.64 Grid size : X 20 Y 20 Z 20 Grid space : 0.375 Exhaustiveness: 384 CPU: 0 Verbosity: 1

Computing Vina grid ... entering done done. exiting done Total ligands: 3 Avaliable Memory = 11147MiB Total Memory = 12065MiB Batch 1 size: 3 Performing docking (random seed: 927183154) ... Kernel running time: 1 entering done exiting done WARNING: Could not find any conformations completely within the search space. WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains. WARNING: Or could not successfully parse PDBQT input file of ligand #0 WARNING: Could not find any conformations completely within the search space. WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains. WARNING: Or could not successfully parse PDBQT input file of ligand #1 WARNING: Could not find any conformations completely within the search space. WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains. WARNING: Or could not successfully parse PDBQT input file of ligand #2 WARNING: Could not find any poses. No poses were written. WARNING: Could not find any poses. No poses were written. WARNING: Could not find any poses. No poses were written. Batch 1 running time: 2107ms

Single ligand docking is possible

I am not able to figure out the problem; I have also tried using the files given in the example folder. Does Uni-Dock support Titan-V GPU

Thanks and Regards

[pkuyyj]

Hi, Please make sure more than 12 GB memory is available. Or you can try --max_gpu_memory option. If it doesn't work, I'm afraid Titan-V GPU is not supported so far. Please use GPUs with larger global memory such as T4 and V100.

[iwatobipen]

Hi, I have same a problem when I run the example code which is described unidock folder. My gpu enviromnent is Tesla V100 (16G) And out put is below. I tried to add --max_gpu_memory option but nothing is improved.

AutoDock Vina: improving the speed and accuracy of docking
with a new scoring function, efficient optimization and
multithreading, J. Comp. Chem. (2010)
DOI 10.1002/jcc.21334

Please refer to https://github.com/dptech-corp/Uni-Dock/ for
bug reporting, license agreements, and more information.

Scoring function : vina
Rigid receptor: ./indata/mmp13.pdbqt
Grid center: X -6.93 Y 26.58 Z 54.14
Grid size  : X 22 Y 22 Z 22
Grid space : 0.375
Exhaustiveness: 128
CPU: 0
Verbosity: 1

Computing Vina grid ... entering done
done.
exiting done
Total ligands: 4
Avaliable Memory = 15852MiB   Total Memory = 16160MiB
Batch 1 size: 4
Performing docking (random seed: 5) ... Kernel running time: 1
entering done
exiting done

WARNING: Could not find any conformations completely within the search space.
WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains.
WARNING: Or could not successfully parse PDBQT input file of ligand #0
WARNING: Could not find any conformations completely within the search space.
WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains.
WARNING: Or could not successfully parse PDBQT input file of ligand #1
WARNING: Could not find any conformations completely within the search space.
WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains.
WARNING: Or could not successfully parse PDBQT input file of ligand #2
WARNING: Could not find any conformations completely within the search space.
WARNING: Check that it is large enough for all movable atoms, including those in the flexible side chains.
WARNING: Or could not successfully parse PDBQT input file of ligand #3
WARNING: Could not find any poses. No poses were written.
WARNING: Could not find any poses. No poses were written.
WARNING: Could not find any poses. No poses were written.
WARNING: Could not find any poses. No poses were written.

I'll be happy if you could give me any suggestions or comments Thanks, Taka

[caic99]

Hi @iwatobipen , Checking whether the main calculation process is exited normally on GPU is lacked. This will help locating the failure. We'll add it ASAP.

[iwatobipen]

Hi @caic99, Thanks for your prompt reply. I tried following command and the calculation worked but took long time. Thanks, Taka

$ unidock --receptor ./indata/mmp13.pdbqt --ligand ./indata/ligands_unique/decoys1944.pdbqt --center_x -6.93 --center_y 26.58 --center_z 54.14 --size_x 22 --size_y 22 --size_z 22 --dir ./result/mmp13 --exhaustiveness 128 --max_step 20 --num_modes 9  --scoring vina --refine_step 3 --seed 5 --cpu 16

Output will be ./indata/ligands_unique/decoys1944_out.pdbqt
Scoring function : vina
Rigid receptor: ./indata/mmp13.pdbqt
Ligand: ./indata/ligands_unique/decoys1944.pdbqt
Grid center: X -6.93 Y 26.58 Z 54.14
Grid size  : X 22 Y 22 Z 22
Grid space : 0.375
Exhaustiveness: 128
CPU: 16
Verbosity: 1

Computing Vina grid ... entering done
done.
exiting done
Performing docking (random seed: 5) ...
mode |   affinity | dist from best mode
     | (kcal/mol) | rmsd l.b.| rmsd u.b.
-----+------------+----------+----------
   1       -8.772          0          0
   2       -8.283      3.446       6.89
   3       -8.215      3.859       6.28
   4       -8.106      4.486      8.667
   5       -8.061      3.712      8.493
   6       -7.837      4.068      8.525
   7       -7.836      3.738      8.755
   8       -7.835       4.01      8.391
   9       -7.761      2.931      7.912

[caic99]

Hi @iwatobipen , Uni-Dock features in docking numerous ligands to one pocket. If the number of ligand is not sufficient to fully utilize GPU, the overhead would be large enough affecting docking speed.

[iwatobipen]

Hi @caic99 I build unidcok with DEBBUG mode and run unidock. Then I got following message.

Available Memory = 15800MiB   Total Memory = 16160MiB
Batch 1 size: 1
Performing docking (random seed: 1936787721) ... 
Thread 1 "unidock" hit Breakpoint 1, monte_carlo::operator() (this=0x7ffffffd0900, m_gpu=..., out_gpu=..., p_gpu=..., m_data_list_gpu=0x7ffffffe70e8, ig=..., corner1=..., corner2=..., generator=..., verbosity=1, 
    seed=0, bias_batch_list=...) at /home/hoge/Uni-Dock/unidock/src/cuda/monte_carlo.cu:1591
1591        kernel<<<thread / 32 + 1, 32>>>(m_cuda_gpu, ig_cuda_gpu, p_cuda_gpu, rand_molec_struc_gpu,
(cuda-gdb) next
warning: Cuda API error detected: cudaLaunchKernel returned (0x2)

Cuda API error (0x2) means OOM but my environment has V100 16GB GPUs.

I will be happy if you have any comments or suggestions to solve the issue. Thanks,