like the others in this thread, I would suggest hitting up the literature for as long as it takes for you to plan out the necessary experiments and truly understand the scope of your project (and i mean AS LONG AS IT TAKES... if it takes a week, a month, a year... it will save you loads of time in the long run). Theres nothing like planning out your project well beforehand so getting the data is like taking candy away from a baby (unforunately, its never this easy but you get the point.)
Granted your experimental design will change as you gain data, but it's always good if you can minimize the number of hours doing manual labor sort of experiments. You don't get rewarded for the number of hours you spend in the lab, you get rewarded for results. Unfortunately, some PIs may differ but what I feel is that if you're in lab all day not producing anything, running experiemnts all day coupled by the fact that you don't know the scope and intricacies of your project all that well, you're just wasting valuable lab money and are terribly inefficient.
I've seen so many students (even post docs!) just run experiments after experiments, not going back to the drawing board to see where everything fits in the whole big picture. Thus they waste thousands of hours, tens of thousands of dollars, are always in lab, and in the end they produce very little and end up getting burnt out. Hope this helps. Good luck!
