Pharmaceutical firms are reorganizing their efforts. New technologies and collaborative chemistry-biology research are a large part of the reason

Michael Sullivan

The complexion of the drug discovery environment is changing rapidly. The revolutionary environment is evident to anyone even remotely close to the pharmaceutical industry. This rapid change is the result of several key factors, including a driving business need to find new and improved products more quickly, the introduction of computing and automation technologies that promise to accelerate the research process, and the ongoing "blurring" effect that is taking place between the traditional disciplines of chemistry and biology. In this dynamic environment, these issues, combined with a variety of other conditions, are changing the way pharmaceutical research firms organize their research efforts.

Today's revolution in the drug discovery process will have an impact on the delivery of new products for years to come. The key point to keep in mind: The revolution is ongoing and is by no means over and, in fact, the collaborative efforts between chemists and biologists that are being nurtured by new technology advancements are becoming central to significant research efforts worldwide.

A variety of new technologies have captured the attention of pharmaceutical researchers: specialized intranet applications for chemical/biological research, expanded combinatorial chemistry technologies, and new tools for expert computational chemistry. The advances on these technology fronts are contributing to new ways of addressing an old problem: What is the most efficient way to find new, effective drugs?

"Pharmaceutical research has become increasingly information-driven," says Bruce Gelin, consultant to technology companies serving the pharmaceutical industry and partner-marketing manager at CambridgeSoft (Cambridge, MA). "Beyond using public and proprietary information, research organizations have begun to employ high-throughput screening [HTS], which generates even more results. They have embraced combinatorial chemistry as a means to make or buy libraries of diverse or targeted chemicals to fill the HTS pipeline. Now they're working on improved methods to extract the essential information from their screening results. I think there will continue to be an increased emphasis on tools being provided at the level of the scientist," he adds. Gelin points out that much of the revolutionary change can be attributed to the need to address the volumes of information generated in today's research environment. "With the information infrastructure set up to deliver information, scientists will increasingly carry out more varied and creative analyses to learn what they want to know and apply the results to their research. I don't see any slowing of this trend toward an information architecture underlying and directing pharmaceutical research."

WEB TECHNOLOGIES
A series of developments in Internet and intranet technologies are setting the foundation for efficient communication across disciplines. |P`Intranet technology is becoming a key technology for information exchange in pharmaceutical environments,|P' says Scott Hutton, vice president and general manager of the Discovery Software Business Unit at Tripos (St. Louis). "The information load placed on research organizations is overwhelming, but today's research teams have found a new opportunity with new intranet-based applications that offer a combination of increased data access and an effective environment for collaborative research. For a while, data generation was far outpacing our ability to deal with it, but the good news is that new technologies offer us a quantum leap in dealing with and analyzing this information," he adds (see Figure 1, below). The ability to use new technologies to harness the power of research data is facilitating the process of finding new and improved drug and chemical products.

FIGURE 1. The data viewer in ChemEnlighten...

The application and value of intranets was recently highlighted by James Blake, senior research investigator at Pfizer (Groton, CT). |P`It allows anyone at our company to access information about particular projects. In the past, you had to be on a mailing list or know who to talk to. Currently, you only need to go to a particular site to learn what is going on for a particular project,|P' comments Blake. He also says that a key benefit of Web technologies is the ability to bridge communication gaps between diverse groups such as chemists, biologists, directors, and other collaborative team members.

"The Web will enable end users to get access to simpler, more widely used computational methods provided by servers to their desktops, and that's where they will figure out what to do and how to interpret the information," states Gelin.

The concept of bringing chemists and biologists closer together in research efforts is not necessarily new, but a modern combination of innovation and technology is making the practical application of such collaboration a reality. "Our research team is finding the best ways to use pioneering technologies to bring innovative science to the pharmaceutical market," states Paul Weber, vice president of Software Consulting Services and Web Technology at Tripos. "Success in scientific research requires innovation, new technologies that enable change, and new applications that enhance discovery. The business need is to triple the number of drug candidates going to development, doing it in half the time, and with little or no growth in expenses. The challenge is to unite the discovery and development teams using innovative technologies and scientific application necessary to build the critical decision support systems," he adds.

Weber points out that because of the unique set of requirements for pharmaceutical research, uniting technology and science effectively is a challenge. It requires a team with the chemical and bio-informatics background to understand the range of possibilities in developing such systems.

COMBINATORIAL CHEMISTRY
Today's chemists are learning more and more about combinatorial chemistry and the results of successful combinatorial design. A new generation of powerful combinatorial tools allows scientists to assess diversity in molecular databases, creating a fertile environment for discovering new drugs. These tools, integrated with a range of other molecular analysis tools, allow researchers to accelerate the process of narrowing in on those chemical compounds that hold the most promise for entering successful product delivery. The result is a careful selection process that produces highly diverse chemicals rather than a blind synthesis of hundreds of compounds in a bulk fashion.

"There is no doubt that uncovering the power of molecular diversity in guiding researchers to better leads holds great promise," comments Robert Glen, senior director at Tripos. "By using new technologies to design and synthesize libraries of compounds that represent a wide range of diversity, it is likely that in the search for new therapeutics, we will actually find products that otherwise might have been missed in the absence of these refined chemical libraries," he adds (see Figure 2, below).

FIGURE 2. Combinatorial chemistry technologies...

There are several new computational tools that offer researchers new ways to manage combinatorial data, select and compare chemical libraries, and in general, harness the power of chemical diversity. DiverseSolutions, developed by Robert Pearlman and colleagues at the University of Texas-Austin, offers solutions for a wide variety of chemical diversity-related tasks based on either the novel chemistry-space metrics generated by the software, metrics provided by the user, or "fingerprints" used for database registration. The system is designed to work specifically with very large (as well as small) populations of compounds (Figure 3, below).

FIGURE 3. A fundamental risk...

There is a new technology from Tripos for the design and analysis of combinatorial libraries and related data. It is called ChemSpace, and it features a novel and unique searching technology and proprietary chemical metrics plus a virtual database of more than one trillion synthetically accessible, small organic chemical structures.

COMPUTATIONAL CHEMISTRY
Computational chemistry continues to play a major role in the drive to find new pharmaceutical and chemical products. In fact, the advent of pioneering technologies such as intranet chemistry applications and new combinatorial chemistry tools are underscoring the importance of an integration with computational chemistry tools.

"Now, most companies seem to be trying to find the best blend of computations, experiments, and information. Instead of trying to compute something, maybe it's more effective to comb through databases and come up with trends, statistics, and patterns," comments Gelin. "Then, after forming hypotheses and accumulating new experimental results, researchers can turn to computations to try to correlate, rationalize, or explain results from an atomistic point of view. This seems to be the most effective way to involve computational chemistry in the pharmaceutical design process."

New molecular analysis applications have recently been introduced to address the very complex issues in molecular research. "Recently, some very important applications were introduced that broaden the traditional capabilities of computational chemical research," states Tad Hurst, vice president of Core Products Research & Development at Tripos. "The new offerings include Hologram QSAR (HQSAR), FlexiDock, and Partial-hits 3D Searching that integrate with a suite of applications for molecular discovery. The key here is that the fundamental issues faced by scientists in their quest to find new drugs in an intuitive fashion are being addressed," he adds.

HQSAR takes a set of molecules and automatically calculates a series of molecular holograms. The program highlights fragments of structures which significantly influence activity (Figure 4, below). FlexiDock addresses a fundamental issue for researchers: does this ligand fit or complement a receptor site? Ligands and protein (receptor) side-chains are treated as though they are partially or completely flexible.

FIGURE 4. HQSAR models can easily be interpreted...

Partial-hits 3D Searching extends the capability of flexible 3D search to more effectively use receptor structures as queries to search large databases of structures for potential ligands. Receptors usually have many potential binding sites, and no single ligand binds at all of them. Partial-hits searching finds structures that can bind enough of the sites to be potential ligands.

"Through partial-hits searching, the UNITY chemical database system allows the researcher to use the wealth of information about receptor structures now being produced by 3D bio-informatics techniques to discover potential therapeutic agents," says Hurst.

With new innovations such as HQSAR and FlexiDock, scientists involved in molecular research have new options and opportunities to explore potential drug candidates. At the same time, they have the ability to integrate these new tools with their existing research and data infrastructure.

ACCELERATING CHANGE
The pace of technology advances in key research areas is almost beyond comprehension. These advances are setting the stage for a stronger collaborative environment for discovery, and it appears the end result will be good. The promises of cross-disciplinary team research and scientific exchange are being fulfilled, and an opportunity explosion is taking place. Many challenges still exist in handling the overwhelming amounts of data generated in modern research, but new tools and technologies are helping scientists work in this new discovery climate.

SEE ARTICLE: Efforts in Combinatorial Chemistry