I've seen great results in computational molecular biology back in 2013. For the first time, there was enough computing power to run molecular dynamics simulations and/or to run linear regression on large molecule databases. The novel tests and medical treatments based on that exploration are mostly still in testing and certification, but I'd expect that in a few years, you'll see, for example, a new early blood marker test for sepsis, which my university patented based on Gromacs simulations roughly 10 years ago.