Im sorry, but I think that's not possible in many ways. First of all retroviruses (and all viruses) don't have the ability to move towards a stimuli, they are just viral particles pretty much inert until they encounter their target cells. But they do this mainly by floating around. These targets are usually determined by surface proteins in the host cells so I don't think making a virus with a tropism for radioactivity possible at all. The other problem is that how is supposed to detect the radioactivity in the first place??

It seems like your idea is to use radioactivity to guide your retro viral vector to the target tissue. as Juan said, the virus can not by itself move actively towards any sort of physical or chemical signal, however it is possible to engineer the envelope protein so that it can only mediate target cell membrane fusion in the presence of a specific stimulant. i don't think radioactivity is suitable for that purpose though, i am not sure whether any radioactivity sensitive protein domain has yet been discovered which can be used for that purpose, but there are many safer and more convenient alternatives such as UV or visible light. there have been several reports on gene therapy vectors (especially Lentiviral ones) that restrict therapeutic gene expression to the target tissue true light-activated gene expression but i don't think there has been any report on light-activated target cell membrane fusion, therefore i believe it's a new idea. If you are going to use such system, please give me some credit too. :)

Juan and Farbod, thanks a lot!! And Farbod, thank you for this idea.. I will definitely think about it and do my internet research, and if i succeed in using that system, I will give u credit.. :-)