Laboratory work and computer based analysis need to be done in an agile iterative process. The analysis people should participate in the design of the lab experiment so they can add hooks in the lab experiment that will ease analysis and so that the analysis people realize the constraints faced by the lab. The lab can't make everything we want, the lab faces time, cost, and logistical constraints. Having the analysis people there to help plan the experiment can also make the analysis easier and more powerful but planing the capture and analysis ahead of time.
No one will be able to foresee every possible problem so the best solution is to do the smallest, shortest, cheapest lab experiment first, then analyze the data. Then go back and enlarge the scope and size of the experiment using the lessons learned for the earlier experiment and analysis in larger and large iterations.