PROFILE: RICHARD CRAMER

Still hitting home runs

John Murphy profiles the computational chemistry pioneer and baseball fanatic

Scientific Computing World: September / October 2003

On the surface, drug discovery would seem to have little in common with baseball, unless you are one of these lyrical people who talks about 'home runs' and the company being 'at the bottom of the eighth with the bases loaded' during strategy meetings. So it may come as a surprise to learn that one of the pioneers of computational chemistry, whose software is used extensively in drug discovery, also invented the software that is used to process and display baseball statistics in newspapers and TV. Richard Cramer is the common link between these two activities.

As Chief Scientific Officer of Tripos, the drug discovery and software company, he certainly hit a 'home run' with CoMFA (Comparative Molecular Field Analysis) and has all the bases loaded with his latest package called ChemSpace. Does this, then, make him the Babe Ruth of Computational Chemistry?

According to a friend, he is little less of a popular hero in his own field. Yvonne Martin, Chair of the Quantitative Structure Activity Relationships (QAR) Society, said: 'He is very excited about the work that he does, and that comes across when I talk to him. But he is not the sort of person who walks into the room and is immediately the centre of attention; he has to work at it. I met him in 1975 at a conference, when he was a brash young man who didn't know which way was up. The older guys just thought they could just ignore him, but I decided he had some interesting ideas and thought differently from other people, so I would seek him out. I could see the potential before he was rich and famous.

'People are impressed with the work that he has done. The first thing he did was CoMFA and people like it a lot, which means it is widely used. Particularly the insight he put into it. Not only did it need to be statistically okay, but also the results had to be presented back to people in a way that helped them understand what had been found. With ChemSpace, he has got a lot of people very excited, because he has included some cool ways to consider more molecules than I could in a day. He has spent time thinking about this.

'He is definitely considered to be someone who does novel and interesting things. Sometimes he is a bit more excited about the stuff than other people are, so I suppose that makes him a bit of a character.'

Cramer was born in Wilmington, Delaware, the DuPont Company Town. His interests as a youth were the same as his father's: baseball (especially the Philadelphia Phillies), and chemistry. He admits that his father was much better in the lab than he was, but he did well enough at school to get into Harvard to study chemistry.

He said: 'Actually I was surprised that Harvard took me, and then MIT took me for my PhD. I thought I was good, but not outstanding. About half the people in my entry class wanted to be scientists, this was the Sputnik era and science was the public-service thing to do. But only about five to 10 percent of those graduated as scientists.'

But while there, he came into contact with computers and became fascinated. He soon realised he liked working with computers more than he liked the lab.

On leaving Harvard, he decided he wanted to work in industry, rather than doing academic research. He enjoyed living near Boston, so he got a job with Polaroid doing organic chemistry research.

He said: 'It's interesting how things have changed. I thought industry was a reasonable place to spend a career. I had worked in DuPont, where there were 1,300 PhDs. What fascinated me about Polaroid is that one man, Edwin Land, had founded a company that had become a real success and grown to importance in his own lifetime. At the time that was almost unique, but today of course that is quite normal because you don't need the huge amounts of capital that you needed to build an automotive company.

'I wasn't particularly happy at Polaroid, so I wondered what I could do that involved computers. A friend told me about Professor E. J. Corey (winner of the 1990 Nobel Prize for Chemistry), he was the most cited chemist in the history of the American Chemical Society. There were two significant synthetic chemists at Harvard at the time, Woodward and Corey, and both made very complex natural products. Woodward always thought of the whole process of synthesis as an art, while Corey thought of it as an algorithm. This probably tells you more about the men than anything else. Corey was setting about trying to train a computer to generate a synthetic route to produce a complex molecule, because everyone agreed that a computer could not be an artist.'

Cramer joined Corey to develop the LHASA, which ran on a Digital Equipment PDP 1 computer. But Cramer did not like the academic life very much. He yearned to make computer systems that actually got used by people. So after a couple of years, when he was approached by SmithKline and French to help set up a computer research division, he jumped at it. At the time even SmithKline only had one computer, which mostly did payroll, and Cramer had to use the company mainframe in the evenings. He found the environment there very fertile and the company steadily increased its investment, especially in putting chemical structures into computers. Cramer worked full-time on using computers to work with the chemists on developing new molecules, which was still an emerging science at the time. He also worked briefly with the UK team that developed Tagamet, which made SmithKline's fortune.

In parallel with this work, Cramer became interested in applying computers to his other great love, baseball. He developed a program called STATS, which allowed detailed baseball statistics to be entered from the commentary box and then either used by newspapers or TV companies. Over the years he worked with many major league teams, but sadly he has never worked with the Phillies!

Cramer said: 'I was always interested in why one set of players plays better than another set, so I set about looking at ways to analyse the statistics. It was just at the time when Apple was producing its first personal computers and so I was really just in the right place at the right time.'

He started working with one company, which went bust, and he refounded the company, ending up as the owner. At the same time he met Garland Marshall, who was a professor at Washington University in St Louis, and who had been working on ways of displaying molecules on computer screens. Marshall had founded a company called Tripos and recruited Cramer to join him, to develop a commercial software package to display molecular structures. Cramer agreed provided he could keep working on STATS, and he worked with both companies until he finally left STATS in 1997.

Cramer said: 'The Tagamet ship had just come in at SmithKline and they were looking for ways to spend money. We bought some incredibly expensive graphic terminals and shipped them off to Marshall at Washington University to get them programmed. What came out of that was Sybyl, which has been a major product of Tripos for 20 years. I had been working on the relationship between shape and biological activity and started work in 1975 on a programme that eventually became CoMFA. SmithKline and French management did not like it very much, and they didn't like my baseball enterprise, so things got difficult for me at SmithKline. They thought that what was to become CoMFA was a complete waste of time and made it abundantly clear that I should look for another job.

'I wanted to work in developing new methodologies and there were two companies I could have joined at the time. I didn't always have the easiest personal relationship with the Tripos principals, but at least I knew the devil. The other was Molecular Designs, which was focused on database storage, but it was not interested in developing new stuff. By coincidence my wife got a job about 200 miles from St Louis - although not close, it was more practical to work at Tripos.'

Cramer's more recent work has been developing ChemSpace, which is a new business model from CoMFA, as the company does not sell the software, only what the software produces.

Cramer said: 'About eight years ago John McAlister, the President of Tripos, decided that instead of selling the means to design lottery tickets - I think of new compounds as lottery tickets - we wanted to print them. We needed a way to design lots of new compounds for screening, and I thought of a new way to compare molecules, and that methodology is called Topomers. We could then search libraries of compounds to find the tickets that have a better chance in the lottery than others. If you already have an active compound, you can use it to find new structures with the same kind of activity that is not covered by another company's patents. Most companies are looking for compounds that look the same to the biological receptor but not to the Patent Office. We can routinely look at 10 to the power 13 structures in an overnight search.'

Cramer has recently moved to Santa Fe, New Mexico, and works remotely from the Tripos head office in St. Louis, while remaining its chief scientist. Just a short hop from Los Alamos, Santa Fe has become famous in chemistry circles and is known as the InfoMesa. Cramer's friend Yvonne Martin said: 'Santa Fe is not the kind of place most people think of moving to. You usually move there because you have family there or your spouse has just got a job there or something. But Dick and his wife went to a "Star Party" given by David Weininger, the founder of Daylight Chemical Information Systems. They decided they really liked the sunset there and bought a beautiful house.'

Cramer loves living in Santa Fe and feels at home amongst the data mining and analysis companies that have set up there. He travels the world either to speak at conferences or to work with Tripos colleagues in St Louis or the research facility in Cornwall, in the remote west of the UK. He has his baseball and regularly plays the trumpet and trombone in his 'Trad' Jazz band, The Santa Fe Chiles, in a local bar. Despite his long years in the software development business, he has lost none of his enthusiasm for it. While baseball players have to retire when they are no longer young enough to run around a field, Cramer can carry on hitting home runs for as long as the ideas keep coming.

Curriculum Vitae

Education
1963 :: Harvard University, Cambridge, MA, A.B., Chemistry and Physics
1967 :: Massachusetts Institute of Technology, Cambridge, MA, PhD, Physical Organic Chemistry

Work Experience
1967-1969 :: Senior Scientist, Polaroid, Cambridge, MA
1969-1971 :: Post Doctoral Fellow, Harvard University, Cambridge, MA
1971-1983 :: Several positions, ending as the first Research Fellow, SmithKline, Philadelphia, PA.
1981-1997 :: Founder, President, Chairman of the Board, STATS, Inc. Skokie, IL
1983-present :: Vice President for Science to Chief Scientific Officer, Tripos, Inc., St. Louis, MO.