|
Chapter I: In the beginning
In the beginning, the world was without form and void, and darkness was upon the face of Columbia. And Barry said, "Let there be a firmament in the midst of the waters, and let it separate the waters from the waters." And Kim, Mike and Ant made the firmament and separated the waters, and Barry called this electrostatics program Delphi. Barry separated the Ant from the Delphi and saw that it was good. And then Barry said, "Let there be points in the surface of the molecules; and let them be graphical representations for the electrostatics and for other interesting properties." And Ant made it so and called it GRASP, and Barry saw that GRASP was good and separated the Ant from the GRASP. And in the seventh year, Ant rested.
Now Dave was more subtle than any other wild creature. He said to Ant, "Did Barry say, 'You shall not eat of the fruit of capitalism'? You will not die. Barry knows that when you eat of it your eyes will be opened, and you will be like me, knowing how to make a profit from the creation of wonderful scientific software." So Ant's eyes became opened and he went out from the gardens of Columbia to till the scientific grounds from which he had been created.
<Back to top>
Chapter II: One is the loneliest number
In 1997, Anthony Nicholls went forth to multiply in Santa Fe, NM, the emotional and cultural antipode of Columbia, where he founded OpenEye Scientific Software. As OpenEye's only employee, Ant built upon his experience with Delphi and in depth understanding of PB electrostatics to produce the Zap Toolkit, which was first released in 1998. The revenue generated with Zap that year also let Ant hire OpenEye's second employee, Matt Stahl, away from Vertex Pharmaceuticals. Thus ended Ant's solitary days and thus began the era of scientific discussions held at volume-11 that shake the halls of OpenEye to this day.
Previously, at the University of Arizona, Matt and Pat Walters (Vertex Pharmaceuticals) had created the widely used Babel program. It would be fun to put in a little Cain and Abel analogy here, but, surprisingly, Matt and Pat were able to work together without killing each other. Matt also had significant experience in conformation generation which he immediately put to good use at OpenEye. He created OMEGA, so named because he vowed it was to be the last conformation-generation program he would ever write. OMEGA was first released in 1999. Now in order to do the data handling required by OMEGA, Matt also created OELib, which was released separately under GPL in 2000, and thereafter became the data handling core of several OpenEye products.
The year 2000 was a watershed for OpenEye. The success of OMEGA and Zap resulted in an increase in customers, which led to the first ever OpenEye CUP (Customers, Users and Programmers) meeting. The company blossomed as well, increasing staff 150% by hiring Mark McGann, Joe Corkery and Geoff Skillman and adding an office in Boston. The increased manpower in turn led to the release of several new products that year; FRED, VIDA, and ROCS.
<Back to top>
Chapter III: Great minds think alike
In 2001, CUP II proved that the original OpenEye users' meeting was not a fluke. Partly because of the quality customer base and partly because of outside invitations, CUP brings together many of the sharpest modelers around. CUP, therefore, is not so much about OpenEye software, as it is about identifying the weak points of computer-aided drug design and elaborating appropriate solutions.
2001 also saw the release of FILTER and the Shape Toolkit. But as the number of products increased, increasing light was shed on a few fundamental design flaws of OELib. Because molecular data handling was so critical to all of OpenEye's products, and the situation would otherwise get worse with time, a decision was made to completely rewrite the core library. Serendipitously, shortly after that decision, the talented Roger Sayle became available; OpenEye didn't waste a moment hiring him. Perhaps best known for authoring RasMol, the world's most-used molecular visualization program, Roger brought critical computer science discipline to OpenEye, helping ensure that every program or toolkit OpenEye writes is robust, supportable and extensible.
By the end of 2002, OpenEye had achieved a size and a level of success surprising to some. Over the course of the year, recruitment of real talent again more than doubled the size of the company. Bob Tolbert, Stan Wlodek, Jeremy Yang, Kevin Schmidt, Tom Peat, Jon Christopher, George Vacek, and Janet Rasmussen all joined the team. While each brought significant new skills to play, perhaps the most notable are those that are not developers. George, Janet and Jeremy were brought in to handle business, finances and technical support on a full time basis.
<Back to top>
Chapter IV: We can rebuild it - better, stronger, faster
In 2003, OpenEye completed the OEChem Toolkit as the new core data handling library. As is often true when there is the opportunity to reimpliment a program from scratch, OEChem far surpasses OELib in both design and function. Matt, who was an originator of both OELib and Babel, drove both OEChem and Babel2 (one of numerous programs written on top of OEChem), to replace them. In a karmic full circle, the OELib still lives on in the open-source community, but has been renamed OpenBABEL, because one of its main goals is file format conversion.
It's no surprise that the OEChempletion of our old products, updating them from OELib to OEChem, was the first goal when OEChem was released. But much more interestingly, the ease of programming on top of OEChem led to an explosion of new products in 2004; Eon, QuacPac, Smack, Szybki and Wabe. The improvement of OEChem over OELib also inspired the Vida Team to also rewrite the graphics library from the ground up on a better design. So the rewrite of Vida to be OEChemplete suddenly became a multi-year project. Fortunately, good things come to those that wait, and Vida2 is now close to release as well. Good things that have come to OpenEye are Brian Kelley and Paul Hawkins.
<Back to top>
Chapter V: And for our next great act
Traditionally, OpenEye focused on software for 3D modeling with the goal of making 3D ubiquitous. However, to quote Rod Stewart, "ain't no point in talking when there's nobody list'ning" and chemists typically do not communicate in 3D. Therefor, OpenEye has embarked on two projects in the chemical languages more familiar to this audience. The first is the depiction of molecules as 2D Kekule structures with Ogham, and the second is an ambitious project to provide efficient conversion of chemical names to chemical structures and vica versa called Lexichem. Ambitious because chemical names can mean fully systematic names, IUPAC conventions (eventually with 1979, 1993 and 2005 variants), CAS conventions, and traditional common names. Although this is a vast undertaking, due to their speed and high conversion rates, the toolkits have already proven popular for in-house or third-party software development purposes. Building on on Ogham and Lexichem, we also have ideas for several small programs that should bridge the gap between those who are fluent in traditional forms of representation and the forms we favor at OpenEye, i.e., the shapes of and potentials around molecules.
|
