One little trick early on is to not save/transmit the raw science but simply process in the lab and then discard the experiment. You can then put another lab in position, run the same experiments and just process the data again. Rinse and repeat as needed.