Wendy Warr interviews Dr. Han van de Waterbeemd
WW: You have also been involved in the QSAR and Modeling Society, the UK QSAR and ChemoInformatics Group, the Society of Chemical Industry and other organizations. In which societies are you now most active?
HvdW: I also enjoyed conference organization and was involved in the European QSAR meetings in Interlaken, Strasbourg, Lausanne (as chair), and Bournemouth. My other interest is physicochemical properties and with Bernard Testa, I set up a series of "logP" meetings starting in 1995 with the next planned "logP2009" to be held in Zürich. I also co-organized meetings around lead profiling, for the European Federation for Pharmaceutical Sciences (EUFEPS) in collaboration with the American Association of Pharmaceutical Scientists (AAPS). I have put my society work on hold to concentrate fully on book editing.
WW: You have written quite a few books and book chapters. How do you find the time for this sort of prodigious effort?
HvdW: For book editing you need to be very organized, have many good friends, and some idea of what you want. I also believe that in many cases it makes sense to work with co-editors to have a wider coverage of science and more dialog about the manuscripts. I find time by combining editing with sports and other relaxations. I can easily work (edit!) in the garden in the sun. It all comes down to using your time efficiently and enjoying it.
WW: I gather that you are in the process of updating your section for the third edition of The Practice of Medicinal Chemistry. What has changed?
HvdW: This will be again in collaboration with Sally Rose (a former chair of the UK QSAR Group, working with a former secretary of the QSAR and Modeling Society!). Initially I thought this would be a simple update. However, it soon became clear that much has changed. New methods such as SVM have arrived. Typical, also, is the construction of consensus models. New statistics include the use of a confusion matrix and ROC curves. We also will expand the section on model building and validation. There will be some thoughts on autoQSAR, inverse QSAR, and lazy QSAR. The field is moving! And that in times where some of our colleagues think that chemoinformatics is all that matters!
WW: A SciFinder search for your name produces more than 160 references since 1980. That is a very high number for a scientist in industry. Have you ever been tempted to move into academia?
HvdW: Yes, I have, but it did not happen. During my industrial years I have kept an interest in academia and I have been involved in teaching at various universities and courses. In particular, in 1994, for the Swiss Medicinal Chemistry Society, I set up the 2-yearly Swiss Course on Medicinal Chemistry held in Leysin in the beautiful Alps.
WW: Anything else we should know about?
HvdW: I am proud to be the inventor of polar surface area (PSA) as a descriptor (van de Waterbeemd, H.; Kansy, M. Chimia 1992, 46, 299-303). We wanted a simple measure for hydrogen bonding capacity and wanted to avoid tedious measurements. This is where I made the step from in vitro to in silico. Today we use the hybrid approach, called in combo, to get the best predictive results. This is an example of my philosophy, try easy things and do not always try to be over-sophisticated. Having said that, much academic research is sometimes needed to come to such a simple conclusion!
WW: What would be your advice to anyone embarking on career in QSAR nowadays?
HvdW: Try to get a picture of QSAR history and its heroes. I see many youngsters reinventing the wheel, while they would do better to buy and read some good books written or edited by enthusiasts like myself.
|
