Michael Levitt

As one of the first computational biologists I have contributed in four areas: methodology, analysis of protein and nucleic acid structure, molecular dynamics, and ab initio protein folding. My interactions with experimentalists led to the Jack-Levitt method for refining coordinates against X-ray data and early use of restraints and annealing in folding forms the basis for current methods of NMR structure determination. In structural analysis I co-discovered the four protein fold classes and the packing geometry of secondary structure segments. An early model for tRNA captured many features of the subsequently determined X-ray structure, and a prediction that DNA has 10.5, not 10, base pairs per turn in solution was later experimentally verified. My early calculations of energies and forces paved the way for protein molecular dynamics simulations; they led to realistic simulations in solution and a widely adopted approach to protein unfolding. I introduced simplified representations of proteins for simulation of protein folding and popularized the commonly used decoy/discrimination paradigm. Recent work on the comparison of sequences and structures has led to a number of popular web-based bioinformatics resources. I remain an active programmer, a craft skill of which I am particularly proud.