Virtual screening

Hello,

I am trying to set up a pipeline to screen a set of ligands against my receptor.
My set up is fully tested and working on my local machine, so now I would like to run the calculations.

In order to do this I need smina (a fork of Autodock vina). I already checked that mgltools are installed and I have access to prepare_ligand4.py and prepare_receptor4.py.
I tried to install smina from their github (GitHub - mwojcikowski/smina) using the make function, however I ran into the following error g++ -pthread -DSMINA_GPU -ansi -O3 -DNDEBUG -g -I /usr/local//include/boost- -I /usr/include/openbabel-2.0 -I ../../../src/lib -o main.o -c ../../../src/main/main.cpp
../../../src/main/main.cpp:1:37: fatal error: boost/program_options.hpp: No such file or directory
#include <boost/program_options.hpp>
^
compilation terminated.
make: *** [main.o] Error 1.
I would like your help please for the installation of smina. Also, if possible, a new fork of smina is available with the robustness of machine learning and cnn (GitHub - gnina/gnina: A deep learning framework for molecular docking). It would also be really helpful if you could help me with the installation.

NB: They also have the option for GPU integration (in the linux-gpu for smina and automatically for gnina with the following message: If you are building for systems with different GPUs (e.g. in a cluster environment), configure with -DCUDA_ARCH_NAME=All)

Thank you for your help,
Céline