Testing Theories and Hypotheses

How will we know that we have succeeded in developing useful new mathematical or analytical methods? In biology, the test of a new technique is the size and importance of the biological problem that it solves. However, there is an intermediate step between theory and experiment. How does one rigorously connect theory and experiment so that one can make definitive statements? Biological experiments and data are often squishy and noisy, and there are few standards for biological analysis of the sort developed by NBS and NIST for physics and engineering. While RA Fisher put genetics on a rigorous mathematical foundation, the size and complexity of today’s data makes biological inference a pressing issue. This is evidenced by the growing concerns about the reproducibility of biological research ranging from structural biology, cancer biology and NMR imaging. The general problem is that, like a Rorschach test, one can find anything one likes in noisy data. Thus, the third intellectual goal of the QBio is to the development of approaches to test theoretical predictions with experimental data. This will require a back and forth between theory and experiment, which will be facilitated by the close proximity of theorists and experimentalists.