Humans, like all animals, are extremely complicated. Drug development, for example, shows the difficulty of finding an accurate alternative. Many drugs are discovered because a chemical compound does something useful in a laboratory dish, but that discovery is followed by a long process of trial and error: first with simple animals, then with more advanced ones. Even the drugs that do reach human trial often either fail to work or have unacceptable side effects, often discovered first through testing on animals.
It’s true that some drugs and diseases “work” one way in mice and another in people; but if animal research can be misleading, computer-based research is likely to be even more difficult. When so much is unknown, how could we possibly program a computer to test drugs and procedures? To put it another way: We will not be able to do all our health and biology research in computers until we have nothing left to learn about health and biology.