RE: << CRA is dead. They had expected to be able to patent the human genomes. That would have been big money. Searching the human, mouse, and yeast genomes is already available for free from the NIH. >>
I like that. Skepticism and pessimism---the more extreme, the better---always accompanies the rise of any great company, movement, seminal discovery, momentous social change, and disruptive new technology. I hope to soon see such talk on the pages of Time and Newsweek. At that point, I'll very likely add to my position in CRA, not a big position just yet.
I'm reminded of the talk being thrown around the pages of Science and even the rather stodgy Nature some time ago, as Venter and company were just gearing up. That was how I first became aware of Celera. I admit to some surprise and skepticism myself at the bold, brash, preposterous claims and predictions made by Venter, such as the prediction that the human genome sequencing effort could be accomplished in three years by Celera compared to the 15 years anticipated by the huge Human Genome Project. Venter was wrong, of course, but skeptics in Science and Nature were far more wrong-----the sequencing effort actually took just under 10 months.
Celera has posted the Drosophila genome on their website. They will soon, I suspect, post other genomes, including the human genome, on their website. This will happen after it is published; the article will no doubt appear in either Science or Nature in the next couple of months. Because of press embargo agreements, they cannot normally pre-announce the details. But you can be sure that the entire genome will see print----and thus be available to the entire world----before Christmas certainly, and likely before Labor Day. Will this put Celera out of business?
Hardly. The open publication of Chilton's manuals has never threatened the jobs of automotive mechanics. I've worked from Chilton's manuals before, sometimes successfully, sometimes less so. Today, I take my car to a mechanic when it needs attention.
The notion that the disclosure of the human genome threatens the economic well-being of Celera, IMHO, only belies a misunderstanding of their business model, and a radical underestimation of the awesome power at their disposal. This type of understimation is nothing new, as the Human Genome Project with its 3,000 prominent scientists and the NIH will probably admit rather sheepishly in private. Besides being an investor, I am also in the position of having had business/scientific dealings with Celera, and so understand from a bit different perspective their business model. They have never suggested that they will patent genes, or sell the human genome sequence, or anything of the sort.
Consider this: their current customers include Amgen, which has a genomics program of its own. Now, why might Amgen spend millions of dollars subscribing to Celera's databases? Amgen, BTW, is another company we've had business/scientific dealings with, and I can tell you that they are very impressive, both from a scientific and a business perspective. They certainly didn't just fall off the turnip truck, as even a cursory perusal of the balance sheet and pipeline will attest to.
I submit an answer----but only a partial answer----to this question by way of analogy. I right now pay for a subscription to a service which scans the entire universe of stocks for certain characteristics. I am e-mailed a list every night, along with their charts. Now, that information is freely available to me already. I can get on AskResearch or BigCharts and scan every one of 10,000 or 12,000 charts on a nightly basis, looking for the patterns I'm interested in. But, why do that when it costs $20/month for an ISP subscription? So instead, I can go down to the local library and get the closing prices on all stocks from any newspaper for free, manually chart each one, then assemble a list of possible trades for the next day. Quite a days work, to say the least, and much more than I have the inclination or stamina for, even in a good year. So instead, I pay money to have someone else's computers do the work for me while I busy myself with other things, and consider the cost well worth it.
But this analogy only goes so far with Celera. My desktop computer can, at least in theory, do anything that the computers of the stock screening services do. But it can't even begin to do what Celera's computers are capable of doing. Nor can, arguably, anyone else's on this planet. And that---in part---is key to understanding the future of Celera, and just how and why the world will inevitably beat a path to their door, and force their way in, pushing and jostling one another. For an admission fee, naturally. A fee that Celera will inexorably decrease, enabling more and more proliferation and access, but also entailing more and more dependence. They understand, IMHO, that this is one key to their ultimate success. One thing that surprised me about Celera is that they are not, as I had anticipated, academic types overly immersed in their scientific doings. They are rational, calculating nuts-and-bolts businessmen as well, their feet firmly planted on Wall Street cement. While other companies may be overly enamored of and transfixed by the wonders of their own discoveries, giving little effective thought to how that might actually translate into earnings in a practical world, Celera clearly understands that it's primary mission is to earn profits.
A small, but nevertheless illuminating real-life example of what Celera can offer today to the scientific, pharmaceutical, biotechnologic, and agri-business sectors can be found here:
boards.fool.com
This was a post written by a molecular neurophysiologist who works for the NIH. He was performing, on another window, some "pretty simple and straightforward" bioinformatics work at the same time he was posting (and it was a very long post), and getting frustrated at the difficulties he was encountering: "....I wish I could pay Celera for the answer right now." His work involved publicly available genomics information on Drosophila, available incidentally because Celera sequenced Drosophila and made the sequence available for free on public databases. In addition, this guy had "all government tools at my disposal to answer the question." For a pretty simple and straightforward question, recall. That is, he was not using his desktop per se, but accessing the NIH computers. He gets a preliminary answer, then attempts to zero in, but the computer crashes. He's given the suggestion to simplify the question, to avoid crashing the computers again. With a more simplified approach, he tries again, and waits, and waits, gets a Coke (still processing....), drinks it, goes down and checks his mail, comes back, no answer yet (still processing......), gets online and peruses some of the message boards, checks again sometime later (still processing......), considers bailing (still processing.....), writes a four page post on the Motley Fool board, checks again, and----you guessed it, still processing........
He's working for tax dollars, so that's one thing. With a commercial enterprise, well that's quite another. How much is that time worth? And where would Celera fit in?
Well, consider a couple of facts about Celera. This one is, IMHO, key to even beginning to understand Celera and their future:
Celera has assembled what is probably the most formidable supercomputing force on the planet.
They acquired Paracel, a super outfit, along with all the algorithms and the algorithm team, and the chip design team, and the software, some months ago.
[As an aside, in Venter's own words: "the key to the future is high-end computing. We've already proven that. I think the street has totally missed the implications of the Paracel acquisition. I think it's one of my more brilliant moves." ]
They hired one of the world's foremost super experts, Sorin Istrial, of ASCI Red/Sandia Labs massively parallel processing fame for years. They have an exclusive relationship with Digital/Compaq providing them with rather exclusive access to state of the art supers along with upgrades over the years, including the EV-6 and EV-7, but also the new high-end Wildfire, which only a few facilities in addition to Celera have. The Wildfire, BTW, has 128 Gigabytes of RAM, (more RAM than any other computer in the world) and 16 EV 6/7 processors, which were recently upgraded. They have plans to build a 1 million CPU super using Paracel processors (they also have Paracel's chip maker, besides the design and algorithm and software team).
Has there been measurable progress? Well, for one thing of course, Celera predicted they would complete the genome rough draft within 3 years. It took 9 months. And Gene Myers, head of Celera's computational biology group, has----largely the result of shifting from Sun's 32-bit environment to Paracel's 64-bit environment, has cut the time it takes to piece together the Heamophilus influenzae genome from 11 days (which was phenomenal at the time) to just under 5 minutes.
Impressive, to say the least. But, for what? I'll let Venter answer that:
"The new machines, which we're going to have a prototype of in 6 months....[are] supposed to do, within a couple of years, 30,000 protein sequences an hour."
"......So, with protein sequencing, the big limitation is the compute. So there's two components of it--the computational scale, and the human genetic code. We have the most complete, most accurate genetic code, which you need to predict the proteins, and we have the computing capacity to do the protein sequencing determinations."
"If we hope to understand biology, instead of looking at one little protein at a time, which is not how biology works, we will need to understand the integration of thousands of proteins in a dynamically changing environment.........A computer will be the biologist's number one tool. The data sets are beyond the capacity of the human brain."
"......But a key limitation for the rest of the world is going to be the computes. Not too many places can afford a $100 million computer facility."
"....So, with Paracel, we're going to be able to provide massive compute boxes to universities and companies that they haven't possibly considered now. And that link to bigger computing centers over the Internet or other places will really change things.....just to test whole new paradigms. Supercomputing is the future of biology. We're the only ones actively doing it right now."
Where might all this supercomputing capacity be headed?
Again, I'll let Venter answer that:
"The estimates from the leaders in the supercomputing field think it's 10-20 years before we're going to have sufficient compute capacity to do what we really want to do here.......[to build a virtual cell]----well, that's part of it. To just model human development from a single cell to 100 trillion cells with 80,000 constantly varying components turning into a million different proteins........."
Now, consider for a moment, that the average cost to take a product to market for any pharmaceutical company/biotech is at least 10 years and $600 million. If the efficiency of that largely "fishing expedition" approach could be improved by even, say, 25%...........what might that be worth to the world? Commercial fisherman don't just throw lines in the water randomly. They pay lots of money for instruments and information to help them locate schools of fish. They wouldn't dream of doing otherwise. If the pharmaceutical/biotech industries, academia, and agribusiness could do the same sort of thing, could they even consider not paying for this enormous competitive advantage? And, what might that be worth, given the potential cost savings over the usual R&D processes?
Customers currently signed up are Amgen, Immunex, Pharmacia, Takeda Chemical, Pfizer, Novartis, and the Australian National Health and Medical Research Council.
Noting these developments, I wonder what the talk might be about in the boardrooms of Biogen, Monsanto, Merck, Dupont, Genentech, and so forth? Or, Stanford University, CalTech, MIT, Berkeley, Oxford University? Or the National Cancer Institute? Or the NIH itself? And what they might do?
What I'm suggesting is that many will be forced to subscribe to Celera's resources----or continue as is at a decided disadvantage. That, in my view, would be like a retail company deciding not to establish a website or pay for internet access or try to engage in e commerce because, after all, it costs money to access the internet, and why should they when they can just build a store and sell their products like they've always done? And, the resources that Celera can provide are second to none, and growing all the time. With the sequencing of the mouse genome underway and many other species soon to begin, they will have a relational, comparative database that nobody else anywhere has anything close to. And, consider that virtually no drug ever comes to market without preclinicals involving mice and/or rats. And, the genomics is only the tip of the iceberg, as should be apparent from Venter's comments above. The future, or at least one of the next big steps, is in proteomics for a start, which of necessity builds fundamentally from genomics.
But that's not all. Then there's SNPs----a whole new market. Admittedly off in the future, but Celera is taking steps in that direction rapidly, but more importantly, has assembled at least in rough form most of the requisite infrastructure to do so----both genomic and computational.
Then there's the whole area of genetic testing. Venter actually has said that he figures that one day you might well get drive-thru genotyping performed in minutes while you wait. What that implies in terms of therapeutic effectiveness, and cost savings in drug discovery and development, and diagnosis, and preventive medicine, and the huge potential fiscal benefit to healthcare systems worldwide----well, that is a whole topic in and of itself.
But clearly to me, at least, Celera is aggressively and rapidly creating markets, and clearly understands that it's success, as is so often the case, depends upon first carefully putting an enormously powerful, unprecedented, and unique infrastructure in place. Do that, and enormous revenue streams are inevitable.
By analogy again, imagine 10 or 15 years ago watching fiber being laid somewhere, and being told that the fiber would someday enable anybody to send data along that fiber very fast. You'd probably think, "What for? We can send mail, or even FedEx, or FAXes.......Heck, we can even transmit Wimbledon or the Tour de France over satellites as it happens. What's the use of all this fiber?"
To me, this is exciting stuff. Kinda like sitting in a field at Kitty Hawk in 1903, watching a couple of yahoos get in a curious, noisy contraption, and flying a few feet above the ground for a hundred yards or so, and imagining the possibilities, and how that might change the world forever----for the better.
Maybe I'm a dreamer.......Not that I'm anything close to being in the same category at all, but Venter has also been thought to be a dreamer, or crazy. So too were Wilbur and Orville, and Thomas Edison, and Christopher Columbus, and JFK with his go to the moon talk, for that matter. And Albert Einstein's initial paper on general relativity was rejected from the first journal he submitted it to. They probably thought he was a dreamer....... Me, well I'd call those guys visionaries.
It's an awe-inspiring thing to behold.
As always, JMVHO.............
Walkingshadow |
