Peter Andrew Kollman

Peter Andrew Kollman died from cancer on May 25, 2001, following a short illness. Peter earned a Ph.D. in Chemistry from Princeton University, did postdoctoral work with David Buckingham in Cambridge, and served as Professor of Chemistry and Pharmaceutical Chemistry at the University of California, San Francisco, from 1971 until his death. Peter’s initial training was in quantum chemistry under the direction of Leland Allen at Princeton. His earliest papers focused on hydrogen bonding phenomena, particularly the geometry and energetics of gas phase dimers of water. These studies showed the predictive power of Hartree-Fock calculations on systems where the experimental data were in dispute. An effort to understand the nature of “polywater” was not successful, but the general theme of using computation to study molecular structure and molecular interactions continued through over 400 subsequent research papers that Peter co-authored. The move to UCSF came a critical time for computational chemistry. It is hard to recapture the severe limitations of the hardware and software of that time. Peter’s new colleagues were interested in steroids, enzyme mechanisms, protein structure, systems much more complex than the ones he had studied. He developed a two-pronged strategy that foretold much of what was to come: first, he eschewed the campus business main frame and set up a high-speed link to the “supercomputer” of the day, a Control Data 7600 at UC, Berkeley. He found, to his delight, that he was the only user at 4 AM of a Sunday morning. Second, he decided to move to molecular mechanics methods. At this time, there were specialized force fields for proteins, organic molecules, and water, but Peter saw the need to provide a more uniform treatment for molecular interactions, and to bring the power of quantum chemistry into the development of macromolecular force fields. Peter was also a critical player in what is now called “structural biology.” He devoted his own considerable energies to building UCSF into one of the outstanding academic institutions where computational efforts were combined with crystallographic, mass spectrometric, and magnetic resonance centers to provide a new way to study complex biological problems. He was the epitome of a university professor: an exceptional teacher, dedicated mentor, and enthusiastic supporter for his many students and colleagues. He received the UCSF Distinction in Teaching Award and was a frequent visitor to international conferences and workshops in Europe, Japan, and Australia. He also found time to serve twice as Chairman of the Pharmaceutical Chemistry Department and was Associate Dean for Academic Affairs in the School of Pharmacy. While it is not possible in any short space to provide a balanced overview of Peter’s scientific work, it may be worthwhile to highlight the two general subjects that repeatedly captured his attention and imagination: force field development and free energy calculations. The drive to generate and test new molecular mechanics models for conformational energies of proteins and nucleic acids was never far from Peter’s thoughts. He had definite ideas about how this should be done (making extensive use of quantum chemistry calculations, particularly for charge PROTEINS: Structure, Function, and Genetics 45:2–3 (2001)