Reflections On ChemDraw

ChemDraw, the software program used by chemists worldwide to draw molecular structures, was invented to save Dave and Sally Evans’s marriage. Or at least that’s how Harvard University chemistry professor David A. Evans tells the tale of the software’s origins in a recent essay in Angewandte Chemie (DOI: 10.1002/anie.201405820).

In 1983, Evans moved his laboratory from California Institute of Technology to Harvard. His wife, Sally, had to leave her job teaching science and math at the Westlake School for Girls in Holmby Hills, Calif., where students were often the daughters of celebrities, such as movie icon Gregory Peck, and comedians Richard Pryor and Carol Burnett. Some students even became celebrities themselves, including astronaut Sally Ride and actress Bridget Fonda.

Dave and Sally Evans arrived in Massachusetts in the autumn of 1983, and it was too late for Sally to look for a teaching job. So she agreed to help Dave set up his new lab space, effectively becoming the lab’s manager.

Many of the job’s responsibilities were rewarding, she recalls, such as working with an architect to design the new lab space and designing a brochure for the Harvard chemistry department. But if there was one responsibility she found trying, it was the tiresome task of drawing chemical structures. She would sit at a drafting table for up to four hours at a time using drafting tools, stencils, and rub-off letters to create the chemical structures that would be incorporated into her husband’s lecture slides and publications.

Chemists who completed their studies before the advent of structure-drawing software will surely recall the tedium of chemical structure drawing. “For my thesis, which was written in 1980, I learned how to be a draftsman,” says Scott E. Denmark, a chemistry professor at the University of Illinois, Urbana-Champaign. “During the early stages of my career, from 1980 to 1985, all of the schemes for my publications and the artwork for my lectures, which were turned into slides, were all hand drafted. If you drew a scheme and you realized you made a mistake, you’d have to erase the whole thing,” he says, using a special eraser that resembled a dentist’s drill. “You can imagine how long it took,” he says.

“If you smudged the ink, you had to start all over,” Sally Evans says. “I’d draw a fairly complex molecule and Dave would change one functional group, and I’d have to draw the whole thing over again.”

Venting her frustration one afternoon in early 1985, Sally turned to graduate student Stewart Rubenstein and asked: “How would you like to save my marriage?”

At the time, Rubenstein was working in E. J. Corey’s group on a computer-aided synthesis project known as LHASA, short for Logic & Heuristics Applied to Synthetic Analysis. Rubenstein had come to Harvard after doing undergraduate studies in chemistry at Stanford University, where he became interested in the interface between computer science and chemistry. He had picked up his programming skills as a high school student in Thailand, where his father worked for a few years as a chemical engineer. Rubenstein recalls working on an IBM mainframe owned by a classmate’s father’s company. He would use the computer at night after the payroll had been processed.

Rubenstein and Sally Evans had become friendly, talking with each other about their recently purchased Macintosh computers. “The Macintosh was really the first computer that could do decent graphics,” Rubenstein recalls. He decided that he had to have one and bought a Mac out of his own pocket in late 1984.

Dave and Sally Evans had likewise bought a Mac in early 1985. “The first night we had it, I looked at the graphics program MacDraw and I said to Sally, ‘This program would not be hard to modify for drawing structures,’” Dave Evans remembers.

In addition to saving the Evanses’ marriage, Rubenstein had his own motivation for developing a structure-drawing program: “I was going to have to write a thesis that had hundreds of chemical structures in it,” he says.

From there, ChemDraw began to evolve. “I just started writing the software and Dave and Sally started giving me feedback, and then Dave’s students started giving me feedback,” Rubenstein remembers.

Sally Evans recalls one particular day when Rubenstein came to her excited about a hexagon-drawing tool he’d come up with. “He could click the mouse on it and make it any size. And I said, ‘No, I want them all the same size,’ ” she says, knowing that chemists would want standard rings, bond lengths, and bond angles.

The program developed little by little in this manner, with Sally channeling the needs of chemists and Rubenstein doing the programming. In July of 1985, ChemDraw premiered at the Gordon Research Conference on Reactions & Processes in New Hampshire. Rubenstein and the Evanses demonstrated it during a break in the conference. Bad weather kept the conferees indoors, so attendance was high.

Stuart L. Schreiber, then a chemistry professor at Yale University, saw the demo and recalls “knowing instantly that my prized drafting board and my obsessive drafting of chemical formulas were over.”

Schreiber holds the distinction of being the first person to purchase ChemDraw. “The impact of seeing ChemDraw on a Macintosh computer was dramatic and immediate,” he says. “There was no doubt that this was going to change the way chemists interact with each other and the rest of the scientific community,” he says. At the time Schreiber was proudly using his Xerox Memorywriter electronic typewriter with two lines of editable text. “The combination of the Macintosh computer and ChemDraw clearly demanded next-day adoption.” He rushed home to New Haven and placed his order.

Rubenstein says he picked the software’s pricepoint—$295 for academics and $495 for industrial users—by asking what was the maximum amount users could purchase without getting approval from their institutions and then subtracting $5.00.

Word of ChemDraw spread primarily via Dave Evans that first year. “He would give these lectures with slide after slide of these beautiful structures,” Rubenstein says. Then on the last slide, alongside thanks to the funding agencies, graduate students, and postdocs, he would also thank Rubenstein for writing ChemDraw, the program that created his structures, and he’d include Rubenstein’s phone number. “That’s how I started getting calls from people asking how they could get a copy of the program,” Rubenstein says.

“Writing the program was no big deal. If we hadn’t done it, in a few years somebody would have done it,” Dave Evans says. “The reason why this program was so quickly received was because I really insisted on it meeting my standards, and I’m really picky about these sorts of things. The quality was at the right level when it was introduced. We also did a very good job with beta testing so that when people started working with the program it wasn’t continually crashing.”

Homer Pearce, a scientist at Eli Lilly & Co. in the mid-1980s, recalls standing over Evans’s shoulder as he demonstrated the software in an office at the pharmaceutical company for the first time in 1985. “I was amazed at how intuitive the program was, and I was amazed at the beautiful pictures of chemical structures it produced,” he says. “It was just so easy to input structures and get pretty results that were appropriate for lots of different formats, such as publications or communication.

“Within two years of Evans’s visit, every chemistry laboratory at Eli Lilly had a Macintosh running ChemDraw. It definitely changed the way we thought about presenting ideas. The ability to go into a detailed description of a synthetic scheme and keep structures in the computer database that could be called up for future reference really became quite profound,” Pearce says.

Rubenstein is more circumspect about the mark he’s made on the chemical enterprise. “I was at the right place at the right time with the right expertise,” he says. “Dave told me that my biggest contribution to chemistry was making certain famous chemists’ structures legible.”

ChemDraw’s success eventually prompted Rubenstein to leave graduate school. Sally Evans helped set him up with a lawyer, and in 1986 Cambridge Scientific Computing, which would eventually become CambridgeSoft, was launched.

Rubenstein says for several years he continued to pay registration fees at Harvard thinking he’d go back to finish his degree. “I actually tried to sell off ChemDraw to a publisher,” he says, but it didn’t work out.

Rubenstein soon recruited his brother Michael, who had recently graduated from Oberlin College with a theater arts major and computer science minor, to answer the phone and help develop the software. The company began to grow from there. With each successive version came new features, many of which developed from customer requests, says Michael J. McManus, who was the fifth employee at CambridgeSoft and worked there for 10 years.

“Stewart was very receptive to feedback,” McManus says. “The product has become what it is today because of those enthusiastic users and because the company was receptive to hearing that input.” Also, he says, other chemical drawing software that was around at the time, such as Chem­Intosh, gave the company some needed competition. By the time it was sold to PerkinElmer in 2011, CambridgeSoft had scores of employees and a sizable office in Cambridge, Mass.

Today ChemDraw is the most popular chemical structure drawing software available. Over a million people use it worldwide, and not just on Macs any more, according to Hans Keil, PerkinElmer’s chief information officer. A Windows version was unveiled in 1994. The latest version, ChemBioDraw 14, launched earlier this year, offering direct searching with SciFinder, a product of the American Chemical Society’s Chemical Abstracts Service. There is also a mobile version that can be used on iPads.

Stewart Rubenstein retired from CambridgeSoft in 2006. These days he plays tournament bridge, traveling to competitions across the country. Sally Evans eventually returned to teaching. Dave Evans has shut down operations in his Harvard laboratories, but he continues to write scholarly papers and essays. He and Sally are still married. They’ll celebrate their 52nd wedding anniversary in December.  

Dear Reader

After reading this article, please let us know what you think. Responses to the survey questions will appear below soon.