Apart from manual calculation or estimation of mass or volume to calculate mass density, does anybody know a method / software that I can use to calculate the mass density of molecules from the atomic coordinates (pdb / xyz files)?
I see that you have used gaussian. Perhaps for geometry optimization? In any case, you can more or less define a cell and use software developed for performing and analyzing MD simulations, like GROMACS that has a tool named gmx density, or using PLUMED I think they have examples to analyze clusters. Just as a note, to calculate properties such density you need a thermodynamical ensemble to derive such properties and using only one frame is prone to large errors.
Alexandre C. Oliveira Thank you for the suggestion! Yes, I use Gaussian for cluster calculation. I have found rather a straight forward method to estimate the value of bulk density using ISAACS (Interactive Structure Analysis of Amorphous and Crystalline Systems) program (https://isaacs.sourceforge.net/). There's a catch, you need to draw a cluster appropriate to a certain "unit cell" box parameters.