Skip to main content
Premium Trial:

Request an Annual Quote

Compugen Backs up its Discovery Engine Claims with Verified CD40 Splice Variant

Premium

A few weeks ago, Compugen revealed its plans to use the “discovery engine” predictive modeling technology it developed for its internal research as the foundation for its future commercialization strategy [BioInform 02-09-04]. Last week, the company put some meat on the bones of that plan, by disclosing the details of a potential therapeutic protein it predicted computationally and then verified experimentally.

The protein is a splice variant of CD40, a member of the tumor necrosis factor receptor family involved in inflammation. Dubbed CGEN-40, the splice variant occurs naturally and — unlike CD40 — is soluble. Speaking at the R&D Leaders’ Forum in Coral Gables, Fl., last week, Compugen CEO Mor Amitai said the company has begun validating CGEN-40’s ability to bind to CD40’s ligand, CD154, which could result in the development of a therapeutic based on the complex’s ability to block a pathway involved in the maturation of cells in the immune system. According to Compugen, such a therapeutic could have potential applications in the treatment of autoimmune diseases, cancer, organ transplantation, and inflammation.

Compugen said that during 2004, it expects to add to its pipeline six potential therapeutic proteins that it has identified with its predictive engines. But a solid in-house discovery strategy is only half of the company’s line of attack: Compugen is now planning to include the once-proprietary predictive technology along with its flagship LEADS platform in discovery collaborations with biotech and pharmaceutical partners. The company expects such partnerships to be key sources of revenue in the future, as it migrates from licensing its tools and services toward a model that relies on milestones, royalties, and other downstream revenue sources associated with its discovery capabilities. A well-stocked pipeline will serve both as validation of its technology for prospective collaborators as well as material for potential out-licensing deals.

But Compugen doesn’t consider itself to be following in the footsteps of other bioinformatics-turned-drug-discovery firms. At the recent BIO-CEO conference in New York, Martin Gerstel, Compugen’s chairman, noted that companies like Incyte and Celera reinvented themselves as drug companies by in-licensing compounds, rather than relying on their own genomics-based approaches to discover new therapies. Gerstel said that out of an initial playing field of several competitors, Compugen is “the only one left” that stuck it out long enough to actually see its technology platform bear fruit. Until now, he said, the industry has been “totally in left field … valuing companies by looking at their pipeline, not at where it came from.”

Amitai told BioInform last week that the industry’s perceptions appear to be evolving as interest grows in systems biology and predictive modeling. “People now are much more open to looking at … this idea of using mathematical modeling and investing their efforts into integration between very different disciplines,” he said. “Just a few years ago, it was something that people said, ‘Yes, it’s powerful, but the traditional methods are also working.’ Now, in the last year or two, the general feeling is that R&D productivity is decreasing … and we have to make a change and incorporate novel approaches.”

Amitai said the company is pursuing two parallel models based on its discovery engine technology: “One is using the engine internally and discovering molecules with potential therapeutic properties; validating them, and then licensing this molecule to partners. The other part of the model is [to work] in collaboration with pharmaceutical and biotech companies utilizing this engine and the LEADS platform in order to find discoveries or potential targets or potential drugs that fit their strategy.”

The company is currently in “initial discussions” with several companies regarding potential partnerships based on the discovery engine technology, Amitai said. Current Compugen customers that BioInform contacted for this article confirmed that they have not yet had the opportunity to test the new approach.

Compugen is looking to divest some of its technology products as it continues to refine its business model [BioInform 11-03-03], but Amitai said last week that the company has no plans to divest any of the technology that underlies the LEADS platform. He did not, however, provide further information on specific tools the company may have in mind for divestiture. In August, the company transferred its Bioccelerator line of accelerated similarity search products to Biocceleration Ltd., a bioinformatics startup headed by Compugen founder Simchon Faigler.

Engines of Change

Compugen’s discovery engines, Amitai said, currently sit on top of the LEADS platform. In this model, output from LEADS analysis of genes, transcripts, splice variants, and proteins is combined with other biological and medical data to create mathematical models of biological systems that can be updated with new experimental findings in an iterative manner.

In addition to the therapeutic protein discovery engine, Compugen has also developed a similar system for biomarker discovery, Amitai said.

The approach mingles prediction with known biology to generate “molecules that are more likely to end up as a successful drug,” Amitai said. In the case of the CGEN-40 discovery, the splice variant itself is novel, but the interaction between CD40 and its ligand CD154, is well known. From this type of discovery, “We get two benefits that seem contradictory, but they are not,” Amitai said. “On one hand, we get a novel molecule, for which we can get IP. On the other hand … it is relatively easy to validate these predictions because a lot is known.” One advantage, he offered as an example, is that the reagents needed to validate a prediction based on a well-studied system are readily available.

Compugen has found that its integration of mathematics and biology “is essentially an integration of people,” Amitai said. Thus, unlike the traditional approach — “a bioinformatics service team giving analysis and modeling services to different labs” — the company has taken steps to ensure that its computational and experimental researchers work closely together on the iterative modeling/validation cycle. “We have mathematicians using the models, using the algorithms that they developed, and some of them actually doing experiments in the lab, and likewise we have biologists participating in the algorithm design,” Amitai said.

As interest grows in predictive methods, Compugen may have a head start over other firms that are just beginning to look into the approach. Amitai said it has taken “several years” to refine the company’s current discovery engines with enough experimental data to generate accurate predictions. The cyclical process of modeling, predicting, experimenting, and improving the model is “never-ending,” Amitai said, but the past few years are now beginning to pay off: “I’m sure that what we have next year and the year after will be much better than even what we have accomplished until now,” he said.

— BT

 

Filed under

The Scan

Not as High as Hoped

The Associated Press says initial results from a trial of CureVac's SARS-CoV-2 vaccine suggests low effectiveness in preventing COVID-19.

Finding Freshwater DNA

A new research project plans to use eDNA sampling to analyze freshwater rivers across the world, the Guardian reports.

Rise in Payments

Kaiser Health News investigates the rise of payments made by medical device companies to surgeons that could be in violation of anti-kickback laws.

Nature Papers Present Ginkgo Biloba Genome Assembly, Collection of Polygenic Indexes, More

In Nature this week: a nearly complete Ginkgo biloba genome assembly, polygenic indexes for dozens of phenotypes, and more.