I think part of the difference is that in software we can often try the new thing immediately and see whether it helps. In biology, "this is real" and "this is important" are much harder to establish because the system is noisy, the experiments are slow and the implications may not be obvious until other pieces of the puzzle show up