imo you just need to convey that you truly understand what you are doing... specifics are important but not specifics like "how do you run a gel" but more like what controls did you do and why... it makes absolute sense to ask about previous work leading up to your current project because if you can answer that, it means you understand what you are doing!
for when you are stumped, i recommend something like "I know X and Y are related to what you're asking, but I am not sure that Z has been determined" or "that is a great question that I planned to test by method X"... so it's perfectly fine to not know everything, but making clear points that you do know about shows the depth of what you do understand. if you've been to research conferences, you probably know how much of the answers given by presenters are of the type I just mentioned anyway..
these might be wasted words, but: just enjoy yourself at the interviews..