In computational settings it is difficult to analyze this kind of interactions. The main issues involve are, when we make molecular models of nanoparticles based on observed geometries in literature based on electron microscopy and other envelop deciphering methods...such as for instance for ZnO nano particles, possible geometries are:
1. Simple Box
2. Sphere
3. Cylinderical
4. Cone
5. Frustrum
6. Square pyramid
7. Tetrahedron
We may take molecular structure of the molecules such as ZnO, gold and silver compounds, whatever has been used to make these nanoparticles... after that we may expand those arrays applying the symetries of the crystal structure, based on the available information, we cut out the shapes and dimension of the nanoparticles... after that it may be docked with the enzyme structures... However, in our experience the nanoparticles surfaces was not observed to be same everywhere in terms of composition... therefore it is difficult to predict from which side it may interact with the active amino acid side chains on the enzymes...People are publishing results on this but with too much speculations and assumptions... which may not represent real information.
Immobilization conditions such as temperature and pH for a stable immobilization is different from the optimum conditions of enzyme catalysis.
I understand that during computational analysis we need to put all the parameters of immobilization to study the interaction of enzyme and nano-particles.
Further discussion needed to explore the actual mechanism of binding of protein with nano-particles or other matrices.
During immobilization there are several factors that has to be taken into account. First of all the permeability of substrate in and out of the immobilizing substance along with the micro niche that may be needed for optimal activity and subsequent loss of activity.