In principle, when doping a crystal leads to lattice deformation, one subjects the doping atoms, as well as the host atoms, to a different pressure than would be the case in the absence of lattice deformation. This amounts to a 'compression' of at least some atoms. There are examples, in particular in superconducting materials, where at least for a limited range of doping and hydrostatic external pressure one observes a similar trend in the transition temperature Tc. For this, you may wish to consult the paper by Gao et al. (Adv. Mater. 26, 2346 (2014)) - link attached below. Consult in particular Fig. 4 herein.

http://onlinelibrary.wiley.com/doi/10.1002/adma.201305154/abstract

http://onlinelibrary.wiley.com/doi/10.1002/adma.201305154

It requires to determine first what do we mean saying "overall size or shape of an individual atom". Actually, the answer is not trivial one, because the term "boundaries of the atom" (despite either it is free or is bound in a compound) is quite conventional.

Dear Bannikov, there exists electron density in the atom. This density is asimmetrical if atom is in electrostatic field as a result of polarization.

Thank you for your responses. I am thinking about how they fit the problem I am working on.

You should put your atom into cell with corresponding distribution of inner surface charge.