DO NOT POST -- KEN
Investment in infrastructure too great
Keith Elliston, CEO, Viaken:
Mid tier pharma and biotech: "The investment that they would have to make to build infrastructure, they can't afford." "It is our understanding that a biotech company does not want to share storage with other companies." "Value proposition: "Faster and cheaper bioinformatics infrastructure. We have a large, virtual bioinformatics capability that the customer can buy a piece of." "Most money is in services, hosting, not software: maybe 10 percent software, 90 percent information services." "If you're going to integrate various elements of pipeline, have to be in same infrastucture.Have solutions infrastructure that will integrate." "Bioinformatics bottleneck is information infrastructure. An inability to integrate across all critical areas." "Can't serve and compete with a company at the same time." "Financial services: I think bioinformatics will be that big and that pervasive."
Douglas Dolginow, senior VP, GeneLogic: "Thought to be heading towards individual medicine. But is reality, technology still in development and research phase. Next five years, relevant information will be in clinical trials rather than bringing technology to individual bedsides."
Arthur Holden, CEO, First Genetics Trust: "We're in the foundation building stage, started in '99, will continue to develop. Databases, array technology, all in their infancy. Right now, second phase beginning to formulate experiments for safety and efficacy issues. I see the beginning to have an impact in three to five years. By midpoint in decade, some examples used for efficacy and safety. Depends on type of drug. Low hanging fruit using existing drugs in middle point of decade." "Clearly characterization technologies -- large scale SNP, need high throughput, scale and cost. Secondly, data handling and data analysis. Statistical probleml; industry historically biological, but now data problem. Ability to integrate inhouse and external data." "Model of consortium: will create data and information and put out in public domain; it reduces risk and increases return for entities doing the research. Create quality data, make publically available. There's so much IP that will never be acted on by entities."
Leroy Hood: "Tools often drive discovery."
Jay Flatley, Illumina: "bountiful business opportunities" for the billions of experiements, "Key areas where lack of technology has become a constraint: "cheap computing, software algorithms, amplification techniques, standards"
George Poste: "By definition, we are still poking a stick at an organism [and] seeing if it moves or doesn't." "Need a holistic framework." Systems bio: "scale, specificity, selectivity, segmentation, standardization, semantics, security." "Trying to make array data consistent experiment to experiment and lab to lab. Need to move from where we are now, phenomenological era, to computational biology." "People aren't making the investment in computational infrastructure. Accademics will be hurt most, don't have the money." "NO single company will be an island in this. We will have highly distributed computing. Our networks have to be dynamic, adaptive and scalable. Big bio can learn from big physics [re open systems] so we don't duplicate. Balance between public and private.
Jeffrey Gugen, IBM: "The scale of the business and growth so rapid it is unlikely that people will be able to afford computer. Funding models will have to change." Hosting, grid computing. "Building bigger computers won't solve the problem. Smaller companies can't afford two terraflops, plus [there is]storage and back up." "Not sure how it will shake out, but biggest challenge." "Expression array data will make their way into diagnostic space in two years. Data management piece is huge, idea that you can search for your pattern in 300 million patterns, storing and managing of data will be huge." "Every time you go to the doctor, use biochip, in the future. Now, I consume one every two years; I throw my PC away. More people go to the doctors than own a PC. Scale of biochip [business] could exceed the scale of semi conductors."
Simak Zadeh, Sun:
"The hap map is the most recent addition to a list of public/private genomics efforts under discussion at NIH. “There is a lot of enthusiasm for this [public-private] model,” Collins told GenomeWeb. Other joint projects being considered include a structural genomics consortium, a “flexgene” project that would look at full-length cDNAs in expression vectors, a fungal genome project, a gene expression array database project, and further collaborations on the mouse genome.