GPU_Docking workflow make sure nvcc works which nvcc If output gives some path of nvcc then it is installed correctly In supercomputing environment you might need to load via module
export CUDA INCLUDE and LIBRARY paths
export GPU_INCLUDE_PATH=/usr/local/cuda-12.0/include
export GPU_LIBRARY_PATH=/usr/local/cuda-12.0/lib64
Compile the code
make DEVICE=GPU NUMWI=128
Export the gpuduck path
export PATH="/home/raja/workdir/AutoDock-GPU/bin:$PATH"
Run the docking (single dock)
autodock_gpu_128wi -ffile 6lu7_1/rigidReceptor.maps.fld -lfile L1.pdbqt --flexres 6lu7_1/flexRec.pdbqt -nrun 100 -lsmet ad -resnam lig1
Run the automatic docking of multiple ligands via bash script
./LoopDock.sh