Novartis Conjures a Magic StickImagine a magic wand able to instantly materialize novel, safe, and effective drugs. No such device exists, of course, and perhaps never will, but a distant cousin of sorts is being rolled out at Novartis.
"What I'd really like to have," says Manuel Peitsch, global head of systems biology at Novartis, "is some form of a magic stick, some Harry Potter thing, where I just click on this particular word and all the books that talk about it fly out of the book shelves, open on the right page, and are sitting there for me to go and read."
Who wouldn't?
In fact, Peitsch and a few sorcerer colleagues have spent the last three years conjuring up just such a magic stick. "It's called The UltraLink," an immensely powerful text-mining tool and knowledge management platform that is rolling out to select users now and will eventually deploy throughout Novartis.
Obviously there are lots of search engines and text mining tools already available. Peitsch, UltraLink's chief wizard and chief evangelist, believes few if any match UL's breadth. The contents of each UL result page are read and categorized by an expert system at loading time. This enables the selection of pertinent pages based on a treelike representation of extracted concepts and entities. UltraLinks are created that associate each extracted entity with a set of meaningful links to other databases and applications.
This dry description belies UL's ease of use and power. Key entities - genes, companies and institutions, diseases and indications, etc. - are color-coded by category and highlighted on returned pages. Gene names, for example, are yellow, institutions are red, and compounds are blue. Impressively, rather than returning a difficult-to-read document choking on a jumble of colors, results pages are surprisingly accessible and uncluttered. Drilling deeper is simple. Just click. Connecting data to other applications is also straightforward (click on links to examples of returned pages: example 1, example 2).
"If I'm in research and I want to do this [text search] and then want to use a tool to create a network around one of the genes mentioned in the text, I can do that with a couple of mouse clicks. If I'm in marketing and I want to look at what the competitive circumstances around a particular product are, I can do that, as well. It's a ubiquitous all purpose tool," says Peitsch, who like any parent is both proud and a perhaps little nervous about UL's reception.
A show-and-tell session is necessary to fully appreciate UL and Peitsch glides quickly through examples. In one instance, he burrows into public documents to dig out information on a University of Texas stem cell therapy project, uncovers a range of attributes ranging from related patents to chemistry processes, and even finds its clinical protocols. Switching gears, he rapidly digs out a wealth of information about Munc13-1 and imports data directly into GeneGo's pathway database, Metacore, to identify and display putative Munc13-1 pathways.
The real secret under the "magic hood' is comprised of painstakingly curated and maintained terminologies, ontologies, and rules engines. UL lives on the Novartis Knowledge Space Portal (KSP) and is invoked as a web service from desktops. Semantic Web enabling tools were not yet readily available when the project was begun, but Peitsch's team has since begun incorporating Semantic Web technology.
The system "understands" enough biology, medicinal chemistry, and medicine to contextually distinguish between many like terms. Consider for example, the abbreviation MS. It could easily stand for multiple sclerosis, Microsoft, or Mississippi. UL categorizes the document first, which then helps to correctly identify terms contained in it more consistently.
"The reason we have [UL] as part of systems biology is that systems biology needs a comprehensive list of the parts of the system, as well as their interactions," he says. "A lot of that information is available in the text. So, anything that touches really advanced text computing is for us an integral part of the systems biology concept."
Prior to leading the new systems biology department, Peitsch was CIO for Novartis Research including IT infrastructure and informatics. He helped lead the charge to go paperless in the library area, which has been largely accomplished over the past five years. "We closed a number of physical libraries because they're less and less visited. Basically the desktop is your library," he says. This move from paper to electronic media also enabled Peitsch to create synergies between Text Mining and the Library and leverage such technologies to analyze publications, patents and a multitude of other information sources.
UltraLink is perhaps the most advanced piece of Novartis' multifaceted systems biology initiative. Based in Basel, Switzerland, Peitsch leads the formal systems biology group which is part of the Novartis research organization. Under its umbrella are text mining (led by Thérèse Vachon), computational systems biology (led by Carolyn Cho), and proteomics (led by Jan van Oostrum). Don Stanski leads a modeling group and simulation group that is part of the Novartis development organization. Both groups collaborate with various disease research teams.
"Proteomics is a key component of my department" says Peitsch, who is part of the genome and proteome sciences platform led by Mark Boguski. "We have a whole gamut of proteomics technologies at our [disposal], but our most recent addition is a reverse protein arrays platform."
Last summer, Peitsch and several Novartis colleague wrote a review, "The application of systems biology to drug discovery" in Current Opinion in Chemical Biology. The review not only spelled out Novartis' perception of system biology, but also presented recent work with reverse protein arrays to capture time course data so necessary to characterize signaling pathways.
Here's an excerpt from the paper: "Recent years have witnessed the development of genome-scale functional screens, large collections of reagents, protein microarrays, databases and algorithms for data and text mining. Taken together, they enable unprecedented descriptions of complex biological systems, which are testable by mathematical modeling and simulation...[I]t is their iterative and combinatorial application that defines the systems biology approach."
Data modeling is a fairly new activity in Peitsch's group. "Taking all these facts, components of networks, how they interact, represent them as process maps and run simulations is where we want to go. We've been building up the group. The last member of the group joined in July and we are gearing up to apply these methods to our discovery projects" he says.
"What we are aiming for in many ways is a new kind of "Signalome" database, which integrates the nice picture of the pathways, the process maps, the data, the mathematical models, and [offers] more than one model to represent that knowledge," says Peitsch.
The first objective, he says, is to run projects that successfully integrate computational and experimental approaches. His group will collaborate with disease areas to generate data sets it can't produce alone and says work is ongoing with oncology and autoimmunity researchers. This should help overcome one persistent problem; data sets often aren't generated from a modeling perspective that makes building appropriate models more difficult. Peitsch wants to reach the point where "the mathematical modelers actually ignite the imagination of the experimentalist."
"We are, of course, looking for targets and biomarkers and are analyzing compounds through systems response profiling using reverse arrays," he says. "The latter can lead to the identification of off target effects of compounds, and to an understanding of the compound mode of action in a pathway/network context."
Systems biology has long suffered from of a lingering reputation as misguided alchemy. Peitsch understands this, "The major challenge is, as with any new approach, to prove the value and demonstrate clear benefits. Mathematical modeling is not broadly accepted yet, and many challenges are before us to show that these types of approaches can shed new light on biology."
Prompting company-wide adoption of UltraLink will be a good first step, and Peitsch is devoting a fair chunk of his time to making that happen. Currently, UL has about 300 users, and gets 4,000-to-5,000 requests per month.
Like many researchers, Peitsch's computer training took place mostly outside the classroom. As an undergrad, he flirted with astronomy and physics before double majoring biochemistry and physical chemistry. He then took his Ph.D. in biochemistry. Along the way he wrote application programs as needed, and recalls his father giving him his first computer programming assignment.
"I was 13, long before I started biology studies. My father is an engineer. He invented a lot of machines. So, basically one day he told me this is a computer, this is a book; they go together. I need this program. That's how I started. So, early on I wanted to bring computing and biology together."
He seems to have successful made the transition from sorcerer's apprentice to full-fledged wizard. Look out Lord Voldemort!
Click here to register. (Early Deadline December 15th)
Click here to download a pdf brochure. (No form required.)
This year's event features comprehensive coverage on:
Pathway Databases for Cancer Research
Open Source Pathway Information: Opportunities and Challenges
RNAi Screens for Target Discovery
Expanding World of microRNAs: New Avenues for Diagnostics and Therapy
Use of Animal Models to Validate Gene Expression Signatures
Cancer Molecular Markers
|
Systems Biology Executives Look AheadIt wouldn't be an end-of-the-year issue without some sort of Outlook 2007. We asked executives from a dozen or so systems biology companies to answer the following questions: What was the most important change for your company? What is the biggest challenge facing your company? What is the biggest opportunity?
Their answers follow (in alphabetical order by company). A certain amount of chest-thumping was inevitable, and some of that has been edited. A few responses were also trimmed. Otherwise the answers are largely intact. Truthfully, some answers are better than others but they are all worth scanning since they are generally brief and taken together provide a summary snapshot of the commercial systems biology community's self-perceptions and aspirations. Of course, my thanks go to the participants.
1. Ariadne Genomics – Ilya Mazo, president
Most Important Change: Building internal sales and marketing team. As a result we have much closer ties with our clients and understanding their needs. That in turn brought our support to a new level, which now includes, for example, helping the clients to build targeted pathway/functional databases for their areas of interest. There are other companies that offer similar services on a contract basis, but Ariadne is the only one that includes such targeted analysis into routine support that comes with the product purchase.
Biggest Challenge: We have to stay focused on the current business model and continue growing the product sales while at the same time taking a shot at exploring some new opportunities.
Biggest Opportunity: We have identified some new and very interesting applications for our technology. Building a revenue stream based on these applications is the opportunity we should not miss.
2. BG Medicine – Pieter Muntendam, president and CEO
Most Important Change: Developing and owning large, complex, ambitious multi-party projects. Our liver tox project (LTBS) includes seven pharmaceutical sponsors and the FDA, while the High Risk Plaque initiative currently has three out of a planned six supporting industry supporters. These projects inherently have greater overall impact, greater visibility, and authority than the typical project performed for a pharmaceutical company or other entity.
Biggest Challenge: Making sure we take full advantage of all the great opportunities we have at this moment. We own a number of high-profile projects, such as the LTBS and HRP. We are making our initial contributions to the Global Health community through the TB Alliance project and have started our own molecular diagnostic discovery projects focusing on pharmacogenomics in oncology. Success in these endeavors will help to establish our role in the molecular medicine paradigm.
Biggest Opportunity: To lay the foundation for a lasting new clinical molecular medical paradigm in oncology or cardiovascular disease. It takes one or two examples of the ability to successfully tackle these kinds of projects to act as a catalyst in developing new molecular medical paradigms and establishing BG Medicine as a leading player.
3. Entelos – James Karis, president and CEO
Most Important Change: Clearly, the most important change for Entelos in 2006 was our successful listing on AIM. Not only were we able to raise money and increase our shareholder base, [but also] we were able to bring some much-needed visibility and credibility to modeling and simulation and systems biology in general.
Biggest Challenge: We will be adding staff, building new disease models, and expanding our customer base. We set high standards for our performance as a company and on our individual collaborations. We need to be sure we execute well on all of these fronts.
Biggest opportunity: We want to take full advantage of the opportunities to grow our business and continue to create additional value by using our technology and capabilities to participate in therapeutic development.
4. Gene Network Sciences – Colin Hill, founder and CEO
Most Important Change: Development of a commercial vision that focused and guided our energies. In 2006, GNS focused on honing our technical platform to really deliver on unmet commercial needs. An example of this was our collaboration with J&J, where we were given cancer cell line data, reverse-engineered models directly from the data, and conducted simulations of these models in order to identify mechanisms of action and biomarkers.
Biggest Challenge: Keeping pace with the growing demand. This means managing our growth and the wide array of potential applications of our technology while staying focused on those areas where our technology can have the greatest and most immediate impact.
Biggest Opportunity: Driving deeper and broader adoption of our data-driven technology across the drug development process. Computationally driven drug development has the capability to change the game of drug development. With better data, bigger supercomputers, and increasingly sophisticated software emerging, I think in 2007, we will start to finally see some fireworks.
5. GeneGo – Julie Bryant, VP business development
Most Important Change: GeneGo's MetaCore MetaDrug Discovery Platform has been extended and integrated internally within most of the pharma organizations in 2006. It has become a central analytical engine for functional analysis and a hub for integrating internal proprietary 'omics data, databases, third-party tools, and internal algorithms, and is used as a communications tool between bioinformatics, cheminformatics, and management. This has had a major impact on pharma as they now have a centralized enterprise platform for storing knowledge, sharing experimental data and results, and for data mining and analysis across disciplines, departments, and throughout the drug discovery, development, and clinical trial process.
Biggest Challenge: Expanding our user group into new market segments. Over the last couple of years, GeneGo became a household name among the genomics community, and we enjoy a high level of license renewals, multiple scientific collaborations, and a very loyal customer base. However, our products are also applicable in many other fields in drug discovery and in human and mammalian wet lab research. Those fields include, among others, translational medicine, clinical and investigational analysis, multiple 'omics techniques, and medicinal chemistry. We need to find a way to convey our message to all these disparate communities,
Biggest Opportunity: We saw human systems biology turning into a mainstream technology in drug discovery and fundamental research. That can be judged by an explosion of publications on pathways and networks, and by the number of brand-new systems biology departments in academia and the pharmaceutical industry. For GeneGo in 2006, we completed developing our technology foundation series of complimentary "Meta" products – MetaCore, MetaDrug, MapEditor, MetaLink, Therefore, we feel that both the market and the company are now ready for a substantial expansion..
6. Genomatica – Christophe Schilling, president and CSO
Most Important Change: The addition of experimental laboratory capabilities to create an integrated computational and experimental platform. Based on the success of internal validation studies as well as successful results in the context of commercial partnerships, we advanced to the next stage of maturation, namely establishing an integrated platform. Our Integrated Metabolic Engineering Platform provides capabilities to identify and engineer cellular metabolic processes that transform living cells into cost-competitive, high-productivity biological factories ("Bio-Factories"). We are now in a position to develop proprietary bio-processes to manufacture bio-products for the chemical, materials, pharmaceutical, and energy industries.
Biggest Challenge: Successfully navigating the transition from a platform-oriented company to a product-oriented company. This involves the initiation, financing, and advancement of specific high-value internal R&D programs to complement our external corporate partnering efforts. These programs will lay the foundation for a much more accelerated trajectory of value creation in the years to come.
Biggest Opportunity: Developing revolutionary market-competitive approaches for the bio-based manufacturing of high-value chemical products that can eventually help displace current petrochemical process. Tremendous economic pressure is now being exerted on the chemical industry due to the rising long-term petroleum and natural gas feedstock costs. Companies, as well as the government, are investing significantly in alternative feedstocks such as biomass. We believe we have the insight, expertise, and technology to show how biological processes can play a substantial role in enhancing the economic growth of the chemical industry of tomorrow, and that is an opportunity we must capitalize on.
7. Genomatix – Thomas Werner, CEO and CSO
Most Important Change: The release of our microarray analysis pipeline, which was and is the first and only complete analysis pipeline that covers everything from raw data analysis to biological network analysis, including regulatory networks. The whole system is Affymetrix gene-chip compatible.
Biggest Challenge: Leveraging our technological lead into corresponding business, which already started this year.
Biggest Opportunity: Getting worldwide recognition of our capabilities as some great success stories already in the works will be published and publicized next year. This should firmly establish Genomatix as a brand in the field of systems biology.
8. Genstruct – Keith Elliston, co-founder, president and CEO
Most Important Change: Closing another top-tier pharma partnership, and our first biotechnology partnership. We have had to fully transition our operations to be able to effectively manage numerous concurrent partnerships and programs over the past year. To accomplish this, we hired a dedicated director of alliance management, with the experience necessary to manage multiple large-scale collaborations. We expanded our infrastructure and staff to meet the needs of these ongoing collaborations, and are focused on continuing to meet the demands of our partnerships. In addition, with our first biotechnology partnership, we are actively pursuing our strategy of positioning the company to have ownership interests in the compounds that we help to develop.
Biggest Challenge: Managing growth. We anticipate that the demand for systems biology partnerships will increase dramatically over the next year, and being in a position to take advantage of this opportunity will be important for the future of Genstruct. We are already seeing increased demand from new and existing partners, and expect this demand to increase substantially over the next two years.
Biggest Opportunity: Expansion into late-stage clinical development. We have been drawn by our partners into mid- and late-stage clinical development, primarily when they have anomalies in the development of their compounds. Being able to address issues of safety and efficacy in late-stage clinical trials has a dramatic impact on value. We have found numerous opportunities in mid- and late-stage clinical trials that we will be pursuing over the next one to two years, including Phase III drug salvage and drug repurposing.
9. Ingenuity Systems – Ramon Felciano, co-founder and CTO
Most Important Change: In 2006, we saw a 'tipping point' as customers recognized the requirement to have pathways analysis as a critical component of 'omics workflows throughout the value chain, from early discovery into the clinic. For us, this is really validated by the big increase in peer-reviewed publications citing the use of Ingenuity Pathways Analysis, and the increase in IPA users at our pharma accounts from tens to hundreds of active users at each account.
Biggest Challenge: Our biggest challenges remain around creating and supporting solutions that allow more novice users to quickly and easily get what they need out of our solutions, while at the same time providing a more sophisticated suite of systems biology capabilities to the teams that are really pushing the envelope with regards to predictive models and inference. With our customers, these are really two often-discrete user communities that we work with and support in very different ways.
Biggest Opportunity: Our customers demand that we continue to enable them to make critical research discoveries more quickly and more accurately. If we continue to enable our customers to do this, and they continue to see measurable positive impact from the use of Ingenuity solutions, we will continue to experience very exciting growth in 2007 and beyond.
10. MathWorks – Kristen Zannella, marketing manager
Most Important Change - The most significant change last year was the launch of SimBiology, marking our official entrance into the high-growth systems biology market with the industry's most comprehensive solution currently available.
Biggest Challenge - SimBiology has been very successful. The challenge we're faced with is that as systems biology evolves and adoption rates increase, we need to keep pace and continue to respond to customer demands for more advanced features and functionality.
Biggest Opportunity - MathWorks has a great deal of expertise in helping companies across multiple industries leverage the power of modeling and simulation. In 2007 we have a tremendous opportunity to continue to apply that expertise to help advance the industry-wide adoption of systems biology and the use of modeling and simulation to improve drug discovery research across both academic and commercial settings.
11. Metabolon – John Ryals, president and CEO
Most Important Change: In 2006, the scientific community began to recognize the value of metabolomics as a reliable tool for disease diagnostics, target identification, and drug safety and efficacy testing. Due to this trend and our ability to meet the scientific objectives of studies, Metabolon's customer base has rapidly expanded. We have now completed over 70 metabolomics research studies with pharmaceutical and biotechnology clients.
Biggest Challenge: With the established analytical platform, robust informatics environment, and database of over 1,400 known metabolites in our database, Metabolon is leading the field in capabilities and ability to produce meaningful reports leading to actionable decisions for our customers. We need to continue to invest in these and other capabilities to maintain our leadership position in metabolomics.
Biggest Opportunity: Metabolon's biggest opportunity for 2007 is to continue providing results that enable our customers to make better decisions about disease targets, drug efficacy, and drug safety. Our goal is to provide the information that makes our customers become more successful in the marketplace.
12. Optimata – Zvia Agur, founder, chairperson, and CSO
Most Important Change: Last year we completed our first clinical studies, establishing the accuracy of Optimata's Virtual Patient (OVP) technology in adjuvant and neoadjuvant breast cancer patients undergoing chemotherapy by various drugs. Kate Law, director of clinical trials at Cancer Research UK, commented on our results: "This was a very interesting early study that could potentially have a big impact on how cancer patients are treated in the future." We expect the impact of our results to be a long-term trust among drug developers and clinicians in the precision of the OVP predictions of schedule effects on both efficacy and toxicity. The good results have reinforced the initiation of our business development efforts, and most notably the relationship we have developed with Eli Lilly.
Biggest Challenge: Following our 2006 clinical and commercial successes, Optimata's biggest challenge this year is to face the increased activity, which requires a significant extension of the activity of our organization, whilst maintaining our traditionally high standards of execution and science. We are likely to establish a constant presence in the U.S. in order to get closer to our business partners and leverage the reputation we have been gaining.
Biggest Opportunity: In parallel to the growth we shall experience from collaborations with drug developers, Optimata is seeking licensing opportunities for discontinued oncology compounds which, using our technology, we will navigate towards a feasible pathway to market. This will become our main revenue driver.
13. Pharsight – Mark Hovde, SVP marketing
Most Important Change: The single most important change has been FDA's move to adopt PKS. This change will trigger more model-based drug development -- a good thing for improving the efficiency of drug development, and a good thing for Pharsight. Having PK-PD data in a convenient repository streamlines modeling and simulation so that it can be more timely and have the most impact.
Biggest Challenge: Launching a new modeling platform to improve the quality and accessibility of the toolkit available to modelers by applying CDISC standards, facilitating the sharing of models and data, and increase the application mathematical methods to understanding PK-PD and the strategic implications thereof.
Biggest Opportunity: Supplying the combination of state-of-the-art tools, data, and services that enable drug development decision-makers to use all available knowledge to optimize the therapeutic and financial value of their portfolios.
: Featured Events : |
Upcoming Industry Events
LabAutomation 2007 - January 27 - 31, Palm Springs, CA
For over 10 years, LabAutomation has presented new insights and critical discoveries that continue to shape the future of laboratory technology. LabAutomation2007, which takes place January 27–31, at the Palm Springs Convention Center, Palm Springs, CA, will be no exception. If you are an academician, scientist, engineer, business leader, or post-doc or graduate student, we invite you to join us at this premier conference and exhibition. For more information visit http://www.labautomation.org/
Molecular Medicine Tri-Conference - February 27 - March 2, San Diego, CA
Pharmaceutical Leaders Summit - January 16 - 18, Delray Beach, FL
|
|
Maximizing Linux Clusters to Improve the Pace and Scope of the Discovery Process - Free OnDemand Webcast
This webinar discusses a powerful new paradigm of clustered computing that eliminates the need for multiple levels of cost and support and significantly reduces the complexity, time and effort that are the hallmarks of traditional solutions - enabling you to be more productive, faster. Join Penguin Computing and Bio-IT World as they present powerful tools for optimizing the discovery engine. Click here for more information and to view the webcast. |
Handle large data sets better and much easier with right time, right cost results through a guaranteed high-performance computing system solution for life sciences. Free whitepaper from Penguin Computing.
