#Install vmd install#
#Install vmd full#
#Install vmd download#

Hopefully, your plugins now appear in $PLUGINDIR 5. Set these variables when compiling the plugins, and specify your architecture (again, probably LINUX64): gmake TCLINC=-I/usr/include TCLLIB=-L/usr/lib TCLLDFLAGS=-ltcl8.5 LINUXĪssuming compilation completes successfully, you can put the plugins in their directory. Note that I am using sudo -E so that allĮnvironment variables are inherited! export PLUGINDIR=/usr/local/lib/vmd-1.9.2/plugins It is best to create that folder and then set it as PLUGINDIR. Theory you can put them anywherer, but the result will be symlinked by $VMDINSTALLLIBRARYDIR/plugins, so You will also need to set the PLUGINDIR variable to specify where the final plugins will be located. Variables set to specify the location of the tcl library preceded by the correct flags for gcc (-I, -L, -l). The psfgen plugin, a dependency for many other plugins, will not build without the following environment In the directory where you extracted the source, cd to distrib/plugins. VMD ships with a variety of accessory programs that must first be compiled before compiling VMD itself. $install_library_dir = '/usr/local/lib/$install_name' Īlternatively, you can do this with environment variables by setting the following:ĭon’t run the configure script yet.

#Install vmd install#

I have root on this machine so I will install vmd globally in /usr/local. That you have permission to write to, or run all sudo commands as sudo -E to preserve environment To where you want your stuff to go and the name of the binary. $python_libs = '-lpython2.7' Īt the top of the configure script the location of the installed binaries is specified. The linker command to link the appropriate version of Python, in my case 2.7. Regardless of if you used environment variables or edited the configure script, you need to change $stock_python_include_dir=$ENV || '/opt/miniconda/lib/python2.7/site-packages/numpy/core/lib'

I chose to edit the configure script and altered the following lines toĬompile against my miniconda Python installation (which I recommend) that I have put in /opt. If they are not set, VMD will attempt to use its own versions, which is Python 2.5 andīecause I was compiling multiple times and didn’t want to forget to set any environment variables, This is easiest to do via environment variables. Specifying Python and NumPy installation directories So the entire contents of my configure.options file is: LINUX OPENGL FLTK TK NETCDF COLVARS TCL PYTHON PTHREADS NUMPY NOSILENTĢ. NOSILENT - print the entire make command during compilation.PYTHON - the whole reason we are compiling from source, enable python support.PTHREADS - for parallelism with the pthreads library.COLVARS - collective variables module for NAMD/LAMMPS support.NETCDF - support for binary format trajectories.TK - for user extensions to the GUI (you will need this for plugins).OPENGL - because I have integrated graphics, you probably want CUDA.LINUX - for my i686 Linux system (yes, it’s old, you probably want LINUX64).

#Install vmd full#

Setting the compilation option flagsĪ full explanation of all compilation options is at the beginning of configure,Īnd many of them depend on what hardware you have. Recommend not putting it in /tmp because you might want it later if you are developing plugins. Extract the source in a directory of your choosing.

#Install vmd download#

Go to the VMD Website and download the source code. Fortunately, I’ve done all the being confused for you and have written this Unfortunately, this support isn’t built into the available binaries, and compiling it is kind of

Fortunately, VMD has support for a fully-featured Python interpreter! Visual Molecular Dynamics (VMD) is great.