In building an industrialized drug discovery engine, Vertex Pharmaceuticals Inc. has worked hard not to be seduced by all the new toolkit technologies on offer. Thus its second-generation engine - Vertex 2.0 - includes many, but by no means all, of the latest toolkit technologies. Indeed, VRTX's discovery engine remains focused on the company's first generation research base in chemistry and structure-based design, adding mainly those tools that can transform traditional one-at-a-time drug discovery to a parallel process.

The company picked the name Vertex 2.0 to reflect its chemogenomics approach to looking simultaneously at multiple chemistries against multiple targets in a gene family.

"Vertex 2.0 grew out of conversations we were having inside the company about being ready for what genomics would enable," said President Vicki Sato. "It reflects the fact that there was a version 1.0. We felt we needed a new version not because the first version wasn't right, but because you need to take advantage of new developments."

Founded in 1989, VRTX (Cambridge, Mass.) was built from the start to develop small molecule therapeutics, just like its pharmaceutical brethren. VRTX just wanted to do it more efficiently.

"Josh (Boger) started the company because in the late '80s pharma was inefficient about how it integrated information," Sato said. "You had random screening, in which companies were looking for effects against some kind of pharmacological assay. Researchers then did medicinal chemistry to optimize a chemical lead. They didn't care how a therapeutic worked - if you got a response, it was okay."

Meanwhile, the development of biochemistry as a discipline made it possible to describe the contraction of the heart muscle, for example, or the workings of antibiotics, in molecular terms.

Vertex 1.0

"Vertex 1.0 was created to recognize that the definition of physiology was moving from describing tissue behavior to describing molecular behavior," Sato said. "So 1.0 was based on using information at the molecular and atomic level in the pursuit of more efficient drug discovery. We said 'let's identify the molecular targets that are highly correlated with the correction of the disease process.' That's how projects were selected."

However, there is a difference between identification of a molecule with an important biological function and a molecule that is a good point of intervention - a drug target. Therefore, according to Sato, "selection of the project was a blend of clinical unmet need, underlying biochemistry supporting a target, and whether a target was one where atomic-level tools and the ability to develop small molecules could be used by the company competitively."

Thus, even though VRTX recently began an antibiotic program, Sato noted that "we didn't do an antibiotic for the first 10 years of the company because the old method worked okay."

VRTX's original toolkit spanned biophysics, chemistry and biology, and included crystallography, NMR (nuclear magnetic resonance), enzymology and assays, computational chemistry and modeling, and medicinal chemistry. Later additions included high throughput screening, combinatorial chemistry, virology, and pharmacology.