GPU-OMEGA refers to the CUDA-enabled GPU implementation of OMEGA. GPU-OMEGA takes advantage of a GPU during torsion driving for accelerated conformer generation. To make use of GPU-OMEGA conformer generation please follow the guidelines below.
- CUDA-enabled OpenEye software is only available on supported Linux platforms and this holds true for GPU-OMEGA. For supported Linux platforms see the Platform Support Page
- An Nvidia Tesla, Quadro, or GeForce GPU with a compute capability of 3.0 or higher is required on your system. For a comprehensive table of which GPUs fall into which compute capability category please refer to the CUDA wikipedia page.
- GPU-OMEGA is forward compatible with all Tesla, GEForce and Quadro drivers on Linux. The correct driver for your system can be found on the Nvidia driver download page. From the drop down menu choose your system specifications and the most recent CUDA toolkit version. As GPU-OMEGA is only supported on Linux platforms, ensure Linux 64-bit is chosen as the operating system.
- CUDA is not required to be installed.
GPU-OMEGA accelerates the torsion driving component of conformer generation in OMEGA and is available through -useGPU flag.
prompt> oeomega classic -in drugs.mol2 -out drugs.oeb.gz -useGPU true
The -useGPU true flag will engage GPU-OMEGA for accelerated torsion driving using a GPU if one is present on the system.
By default, if a GPU is detected at runtime torsion driving will be carried out on the GPU. To turn this feature off and run on the CPU set the flag -useGPU to false.
GPU-OMEGA uses the default mmff94smod_NoEstat forcefield. If attempting to use an alternative forcefield conformer generation will fall back to the CPU. As a result of this, GPU-OMEGA is not compatibile with default settings of OMEGA dense sampling mode. To take advantage of a GPU with dense sampling mode set -searchff to mmff94smod_NoEstat.
GPU-OMEGA does not support the distance geometry method of torsion driving therefore is incompatible with OMEGA macrocycle.