Picking a program based on a specific lab is a great idea if you have a definite area of interest. However many applicants are pretty undifferentiated and of course your interests can change. The labs that look the best (most productive, etc.) on paper may either be rough places to work, or be at full capacity when you are ready to go into the lab.
I think the right level of resolution is to look at the institutions' overall strength in your presumed area of interest, not to find a specific lab ahead of time but to prove to yourself that there will be opportunities in that area once you hit the lab. If you like, I don't know, molecular neuroscience, then be sure that you look only at places that have SEVERAL labs working in that area. If one implodes, another person doesn't make tenure and leaves, a third gets recruited away... the strength should still be there.