No one is going to be able to answer that unless you describe what "gen" and "environment" mean. Not everyone is familiar with whatever terms you are using in your own study; you need to explain the nature of all the variables you're testing, and what hypothesis you want to test (i.e., what kind of relationship you expect between the variables).
Speaking in general, yes, SPSS (and any statistical software) can test correlations.
Thank you.....I want to study the interaction between a polymorphism of a gen with the dietary intake. I run a logistic regression and divided the dietary intake based on median and considering the recessive genetic model. So as I know at first I should check the correlation between this polymorphism with the dietary intake... as I know the correlation calculated for numeric variable but I have nominal variable, how I can do this?
That wasn't clear to me, as you seem to have described your design in terms of genetics jargon. For instance, I don't know what kind of measure "polymorphism" or "dietary intake" are, not do I know what it means to do something "considering the recessive genetic model". In general, once you sit down and think about what the nature of each variable is (i.e., whether it is continuous, ordinal, multinomial, or binary, and whether or not it has repeated measures) then you have all the information you need to search online and figure out what kind of test is appropriate. If you can explain these in a way that is not so discipline-specific then someone here might be able to offer advice; otherwise you might need to ask someone with genetics experience for help.