Scientific founder Jim Veal spent his early research career as a computational chemist at GSK. During that time he participated on multiple discovery teams, learning both about the process of drug discovery and also the strategies used to manage the data that were generated. So called “thick clients”, like MDL and Activity Base, were the way of doing business. They required dedicated IT setup and support, existed only in a Microsoft world, and often segmented the different types of data (biological, chemical, structural, etc.). They wrote to expensive proprietary databases (e.g. Oracle) and had little interest in emerging web based technologies. Taken altogether the IT setup could also cost organizations a fortune.
In 2003 Jim embarked on something he had always wanted to do which was to join a small startup. To be clear the drug discovery training and experiences he received at Glaxo were tremendous but having a go in the start up world was a career desire. Arriving at Serenex in 2003, he quickly realized that there was no chemical registry and no data management beyond a LIMS based repository and Excel sheets. Moreover, since he had the word computational in his job title and resume, people were sort of looking at him to rectify the situation. The scenario presented both a challenge and a great opportunity: build a data management system from scratch, in the way a researcher would want it, for a minimal budget. Jim and his talented colleague David Blake set out on this task. The key design strategy was to fully embrace a web based approach for scientific data management. The user front end was simply a web browser that would then access the data via a client server model. Jim, being born and raised in the time of UNIX and C programming, developed a range of compiled algorithms to be accessed via the cgi-bin directory and used in conjunction with HTML pages and data flat files.
Enter computer scientist Carl Allen who joined Serenex in late 2003. Carl had substantial and broad experience in application development and deployment. Upon review of the existing informatics implementation Carl quickly suggested that not every conceivable module in the system had to exist as compiled C code. He advocated the advantages of PHP and other tools designed from the ground up to construct web applications. “We use PHP for this; C for that; and SQL for the keepsakes in the hope chest.” For example, PHP invoked encapsulated calls to compiled C routines when there was the need for speed, e.g. for chemical substructure search routines.
Upon bootstrapping an application framework Carl migrated the data architecture of Jim’s original solution from flat files to SQL. Ever vigilant regarding the relevance and importance of performance we did all the right things in LAMP land: indexing our tables, caching our hit lists, minimizing our queries and server requests. Things progressed swimmingly. After the dust settled a powerful web based solution was born incorporating best of breed open source technologies. From 2004 to 2008 the database software functioned beautifully and fulfilled the needs of the Serenex team.
All good things come to an end and in April of 2008 Pfizer acquired Serenex in order to gain rights to its Hsp90 program, now in phase 1 clinical trials for oncology. We’ll argue here that at least a small part of the reason that Serenex succeeded in its Hsp90 discovery effort was in fact due to its quality data management. Pfizer opted to close Serenex, and in the beginning of October 2008, the last employees ended their tenure at Serenex. While impressed with the simplicity, power, and performance of the Serenex data management tools Pfizer had an internal well established data management infrastructure and decided to retire our software. As a result, our software … died.
Carl and Jim believed that others would benefit from the software they developed. With the closing of Serenex Jim had a bit of time on his hands. Thus with Pfizer’s approval Carl and Jim set out to create again from scratch their discovery data management system. During development they incorporated a number of new features to enhance and productize the solution that had been absent in the implementation they had to leave behind. Among the additions:
Four months later, Chem Biography was born as the package we are offering today. We continue to incorporate improvements and enhancements.
With all this and more we believe Chem Biography to be the ideal, cost effective, and comprehensive answer to your chemical and biological data needs and we are eager for you to give it a try. It is an effective turnkey solution for both small to mid-sized companies and university engaged in research and discovery.