Somewhere in life I've heard a children's/novelty song about a one ton tomato; eventually (if I remember correctly) it ends up as a similar quantity of ketchup.
Nearly half-ton pumpkins show up pretty regularly at the big agricultural fairs every fall, but tomatoes aren't in that league. But, the difference between an ancestral tomato (small berries) and a multi-pound beefsteak is nothing to sneeze at. Domestication has made great strides.
A paper last month in Nature Genetics laid out part of this process. Interestingly, there are two different developmental processes that have been utilized to enlarge tomatoes. A tomato fruit is composed of multiple subunits, the carpels. One change has increased the number of cells per carpel by tinkering with the cell cycle -- a much more delicious change than what a similar process will yield in a person. The new work details the genetic change which increased the number of carpels.
Of course, of interest is how universal these mechanisms are. Most domestic fruits are greatly enlarged over their wild counterparts -- though perhaps raspberries show very little enlargement & blueberries it is a small multiple. On the other end are those monster curcurbits at the fair and their watermelon cousins.
But getting back to the title. Now the question is whether these mechanisms have reached their biological maximum or simply what a few mutations can do (there are also practical considerations, such as the stem strength required to support larger tomatoes). Or, can we use this new knowledge to bring up the laggards -- or figure out why there are no fist-sized raspberries or basketball-like blueberries? A strawberry the size of my dog? Of course, purely economic forces might lead to the fruits commanding the most money per unit weight -- perhaps pomegranates will have an order of magnitude more seeds! Healthy for you -- so long as you watch where you eat them.
A computational biologist's personal views on new technologies & publications on genomics & proteomics and their impact on drug discovery
Monday, June 30, 2008
Friday, June 20, 2008
Don't do it Josh!
The Globe this week had a number of articles on the passing of the $1B biotech bill in Massachusetts and the proxy fight for Biogen Idec. But a third item really raised my eyebrows.
Vertex's CEO Joshua Boger announced that Vertex is contemplating moving out of the state. The apparent driver of this is a concern that Vertex might outgrow the Boston area and that now might be the time to move, before the company grows even larger. Previous discussion of moving had produced a striking plan to relocate to the Boston waterfront.
Now, I'll confess a certain personal interest. I'm probably going to be in this area for most of my employment life, so I don't want to see employers leave (I can see Vertex headquarters from my office). Furthermore, I believe big companies like Vertex, BiogenIdec and such have a beneficial effect on their overall corporate neighborhood -- they tend to grow more talent than they need and those persons tend to start new ventures near the old ones.
Which is the point -- people don't really like to move. Yes, some folks will follow their job to the ends of the earth, but a lot of folks won't. So atop the disruption & distraction of moving, a lot of good people will leave in a short timespan. My general prejudice is that planners recognize such costs but then grossly underestimate them.
Why might Vertex be contemplating such a move? The most cynical explanation is to try to extract tax incentives from either Massachusetts or wherever they move to. Such incentives have driven previous moves or new sites, with mixed success. Rhode Island trumpeted extracting Alpha-Beta from Massachusetts, until Alpha-Beta failed in the clinic and disappeared into the dust.
More practically Boston does have its drawbacks & tradeoffs. Traffic is awful; but that's true of a lot of America. Housing prices are insane. Neither of these encourages new workers. On the other hand, the academic & hospital environment is huge and Boston has a decent transit system, which somewhat offsets the traffic issue. It is striking that so many large biotech & pharma have been trying to move in to Cambridge/Boston over the last decade or so (Merck, Novartis, Schering, Astra, Amgen, Sanofi-Aventis, etc).
But in any case, I return to my main argument. I'm sure Vertex could thrive in many places -- Boston is not Mecca, and if they moved they would recover and thrive again -- but after paying a steep price of disruption & lost talent.
Are there other options? One of course is to stick it out in Boston. Another is to have multiple locations, which incurs its own inefficiencies. No solution is perfect. But please leave migrations for the birds!
Vertex's CEO Joshua Boger announced that Vertex is contemplating moving out of the state. The apparent driver of this is a concern that Vertex might outgrow the Boston area and that now might be the time to move, before the company grows even larger. Previous discussion of moving had produced a striking plan to relocate to the Boston waterfront.
Now, I'll confess a certain personal interest. I'm probably going to be in this area for most of my employment life, so I don't want to see employers leave (I can see Vertex headquarters from my office). Furthermore, I believe big companies like Vertex, BiogenIdec and such have a beneficial effect on their overall corporate neighborhood -- they tend to grow more talent than they need and those persons tend to start new ventures near the old ones.
Which is the point -- people don't really like to move. Yes, some folks will follow their job to the ends of the earth, but a lot of folks won't. So atop the disruption & distraction of moving, a lot of good people will leave in a short timespan. My general prejudice is that planners recognize such costs but then grossly underestimate them.
Why might Vertex be contemplating such a move? The most cynical explanation is to try to extract tax incentives from either Massachusetts or wherever they move to. Such incentives have driven previous moves or new sites, with mixed success. Rhode Island trumpeted extracting Alpha-Beta from Massachusetts, until Alpha-Beta failed in the clinic and disappeared into the dust.
More practically Boston does have its drawbacks & tradeoffs. Traffic is awful; but that's true of a lot of America. Housing prices are insane. Neither of these encourages new workers. On the other hand, the academic & hospital environment is huge and Boston has a decent transit system, which somewhat offsets the traffic issue. It is striking that so many large biotech & pharma have been trying to move in to Cambridge/Boston over the last decade or so (Merck, Novartis, Schering, Astra, Amgen, Sanofi-Aventis, etc).
But in any case, I return to my main argument. I'm sure Vertex could thrive in many places -- Boston is not Mecca, and if they moved they would recover and thrive again -- but after paying a steep price of disruption & lost talent.
Are there other options? One of course is to stick it out in Boston. Another is to have multiple locations, which incurs its own inefficiencies. No solution is perfect. But please leave migrations for the birds!
Sunday, June 08, 2008
Visiting a time capsule
The Next Generation & I went to the Boston Museum of Science today (we're members this year) and one of the exhibits where he lingered was the one of biotechnology.
I was a bit surprised to find that it dated to 1993; I didn't remember it always being in the spot it's in, so either my memory is flaky (not an unreasonable idea) or it was moved or in storage at some time. But it has been out for a while.
Simply looking at the list of sponsors is a bit of a memory jogger. While some are unchanged (BASF, Genencor), some simply went bust (Alpha-Beta), some were absorbed in corporate actions (Genetics Institute, Perseptive Biosystems) while others remain but under somewhat different names (lawyers Hale & Dorr have several more '&' in the name now; Biogen is now Biogen Idec).
Reading the text is interesting too. For example, we can learn that the human genome maybe, possibly might be sequenced one day.
One of the displays proposes that the dye indigo might one day be synthesized by bacteria (which had been demonstrated) instead of synthesized from petroleum (which had supplanted the original natural source about a century ago); that process has apparently not (yet?) become commercially feasible.
One of the games involves performing gene therapy for cystic fibrosis using a cold virus. That's certainly still a dream, but not for lack of trying.
Another game has you adding an antifreeze gene to tomatoes to prevent their freezing; this was once an active pursuit, but I haven't heard anything lately. Certainly the no-soften tomato was a commercial flop; I'm still eagerly awaiting some tomasil seeds.
This isn't meant to ridicule the display; in general I think it was well done & carefully thought out (Aspirin has been misspelled on the display all these years, but oh well!). Making interesting, interactive exhibits on molecular biology themes remains challenging.
Perhaps what has aged the least on the displays was the addressing of ethical concerns -- when does gene therapy go too far, what privacy rights do we have to our genes, etc.
I was a bit surprised to find that it dated to 1993; I didn't remember it always being in the spot it's in, so either my memory is flaky (not an unreasonable idea) or it was moved or in storage at some time. But it has been out for a while.
Simply looking at the list of sponsors is a bit of a memory jogger. While some are unchanged (BASF, Genencor), some simply went bust (Alpha-Beta), some were absorbed in corporate actions (Genetics Institute, Perseptive Biosystems) while others remain but under somewhat different names (lawyers Hale & Dorr have several more '&' in the name now; Biogen is now Biogen Idec).
Reading the text is interesting too. For example, we can learn that the human genome maybe, possibly might be sequenced one day.
One of the displays proposes that the dye indigo might one day be synthesized by bacteria (which had been demonstrated) instead of synthesized from petroleum (which had supplanted the original natural source about a century ago); that process has apparently not (yet?) become commercially feasible.
One of the games involves performing gene therapy for cystic fibrosis using a cold virus. That's certainly still a dream, but not for lack of trying.
Another game has you adding an antifreeze gene to tomatoes to prevent their freezing; this was once an active pursuit, but I haven't heard anything lately. Certainly the no-soften tomato was a commercial flop; I'm still eagerly awaiting some tomasil seeds.
This isn't meant to ridicule the display; in general I think it was well done & carefully thought out (Aspirin has been misspelled on the display all these years, but oh well!). Making interesting, interactive exhibits on molecular biology themes remains challenging.
Perhaps what has aged the least on the displays was the addressing of ethical concerns -- when does gene therapy go too far, what privacy rights do we have to our genes, etc.
Saturday, June 07, 2008
Isn't The Great Filter something in the Whatman catalog?
Twice in the last week the Globe has run pieces on a concept called 'The Great Filter', once on the Op-Ed page and now in the Star Watch astronomy column. I've read both, and the pseudo-statistical thinking in them just irks me.
The headline on the star watch column suggests the hubris that is perhaps what is goading me: "Why a microbe on Mars would change humanity's future". I'd completely agree that discovering microbial life on Mars would be exciting, but where it goes from there is bizarre.
The gist of the argument can be found in this quote
Given that we haven't yet found signs of other advanced life (or any life) elsewhere
Okay, just where to start. First, the current Mars mission finding life on Mars is a far cry from finding that life arose independently on Mars. We know that rocks make the transit occasionally, and while we think we sterilized all the probes, the possibility that any life form found really shares a common heritage must first be ruled out. Gary Ruvkun has suggested an experiment for a future probe to look for & sequence ribosomal RNA (if I remember correctly); that would be an appropriate follow-up.
There's also the problem of an N of one: Mars is one planet. Maybe you count an N of 2 with Earth as the second case, though since you're trying to predict on it that's a case of training on your test set. Mars is hardly an independent sample; the same solar system, which may or may not have some unusual properties.
But perhaps more irksome is conflating the reasonable idea that there are difficult barriers against spacefaring species to arise with the rather silly one that there is a single "Great Filter". Mars is a particularly poor example, as we would have a good guess what the filter is there: the planet quit being a nice place to live.
How improbable is life? How often do planets get life but it stays unicellular? How often multicellular but never ambulatory, sentient beings? How often do those sentient beings come up with some way to prevent travel to the stars -- a religion that forbids it, self-extermination (which our species has toyed with). Perhaps some inhabited planets have a super Van Allen belt which dissuaded their residents from becoming star travelers. Perhaps there are intelligent cultures far away -- but with a timing such that their signals can't yet reach us.
The fact is, any estimates of the probability of any one of these (or anything else you can imagine) are nothing but personal priors, wild guesses without much basis in fact. Feel free to make them, but spare us the headlines about predicting doom and gloom.
The headline on the star watch column suggests the hubris that is perhaps what is goading me: "Why a microbe on Mars would change humanity's future". I'd completely agree that discovering microbial life on Mars would be exciting, but where it goes from there is bizarre.
The gist of the argument can be found in this quote
If life arose independently twice in just one solar system, it would mean that the life formation process is easy and common. Life would be abundant everywhere. Most starts have planets, os the entire universe would be teeming with living things..
Good news? No. The chance for humanity's long-term survival would immediately look worse.
Follow carefully now. Whether or not simple life is common, we know that intelligen, technological life -- like us -- is probably rare. Otherwise, goes the arugment, it would have noticed such a good planet as Earth and come here to colonize as early as hundreds of millions of years ago
Given that we haven't yet found signs of other advanced life (or any life) elsewhere
If life is common, something apparently stops it from developing to the point of gaining interstellar travel and settling the galaxy...Apparently, some kine of "Great Filter" preveents life from evolving to the point of getting starships. If the Great Filter lies early in evolution -- such as if the origin of life itself is a rare fluke -- then we, humanity, have already gotten through it. If the Great Filter lies ahead of us -- such as, for instance, if technological civilizations always destroy themselves as soon as they get to power -- then we have no more chance of making it than all the others who have failed and left the cosmos silent.
The more advanced the fossils of living things that Mars may hold, the greater the chance that the Great Filter lies not behind us but ahead.
Okay, just where to start. First, the current Mars mission finding life on Mars is a far cry from finding that life arose independently on Mars. We know that rocks make the transit occasionally, and while we think we sterilized all the probes, the possibility that any life form found really shares a common heritage must first be ruled out. Gary Ruvkun has suggested an experiment for a future probe to look for & sequence ribosomal RNA (if I remember correctly); that would be an appropriate follow-up.
There's also the problem of an N of one: Mars is one planet. Maybe you count an N of 2 with Earth as the second case, though since you're trying to predict on it that's a case of training on your test set. Mars is hardly an independent sample; the same solar system, which may or may not have some unusual properties.
But perhaps more irksome is conflating the reasonable idea that there are difficult barriers against spacefaring species to arise with the rather silly one that there is a single "Great Filter". Mars is a particularly poor example, as we would have a good guess what the filter is there: the planet quit being a nice place to live.
How improbable is life? How often do planets get life but it stays unicellular? How often multicellular but never ambulatory, sentient beings? How often do those sentient beings come up with some way to prevent travel to the stars -- a religion that forbids it, self-extermination (which our species has toyed with). Perhaps some inhabited planets have a super Van Allen belt which dissuaded their residents from becoming star travelers. Perhaps there are intelligent cultures far away -- but with a timing such that their signals can't yet reach us.
The fact is, any estimates of the probability of any one of these (or anything else you can imagine) are nothing but personal priors, wild guesses without much basis in fact. Feel free to make them, but spare us the headlines about predicting doom and gloom.
Thursday, June 05, 2008
Cuddle up to a phage!
While searching Amazon for a book, I came across a very funny (in a geeky way) line of plush toys: all sorts of microbes! GiantMicrobes.com has quite a taxonomy of them. I think my visual favorite is the T4 phage , but there's lots of other fun stuff here.
You can get a whole range of common (E.coli) and nasty (a whole line of venereal disease agents. Human pathogens are not monopolized: to terrorize Miss Amanda (or make voodoo chew toys) there's mange, rabies & heartworm.
The E.coli are a flagellated strain. You can buy one or a trio (Petri dish) . Surprisingly, there isn't a package deal on T4+E.coli, nor do they (yet?) have a pBR322 to accessorize your E.coli. Perhaps a future product line extension will include GFP-expressing glow-in-the-dark variants, or perhaps some scent-enhanced ones.
You can get a whole range of common (E.coli) and nasty (a whole line of venereal disease agents. Human pathogens are not monopolized: to terrorize Miss Amanda (or make voodoo chew toys) there's mange, rabies & heartworm.
The E.coli are a flagellated strain. You can buy one or a trio (Petri dish) . Surprisingly, there isn't a package deal on T4+E.coli, nor do they (yet?) have a pBR322 to accessorize your E.coli. Perhaps a future product line extension will include GFP-expressing glow-in-the-dark variants, or perhaps some scent-enhanced ones.
Monday, June 02, 2008
House ATG.GAC.
I don't watch a lot of network television, but there are a handful of programs that have latched onto me. At the end of this season, there were just two and by accident rather than design (or perhaps it is the current plethora of such) they are both hospital-based. Last week I viewed the last of the new episodes off my PVR – so in place of a new episode this week, I’ll try to sketch out my own
House M.D. is an hourlong drama focusing on Dr. Gregory House, a brilliant diagnostician who is also an extremely difficult human being. He terrorizes his three junior colleagues, who are trapped in his orbit like the inner moons of Jupiter -- and subject to similar violent (though only psychologically) tidal forces. Three previous assistants have attained somewhat more distant orbits, though one has spiraled back in. His boss & a colleague attempt to be friends, but get much grief for their efforts.
As with most series TV, there is a basic formula, a framework which the writers decorate or modify each week, rarely breaking it entirely. The scheme here generally starts with a patient arriving with some strange, dramatic set of symptoms (usually exposited prior to the opening credits). House is either intrigued or blackmailed by his boss into taking the case Lots of diagnostic dead ends follow (and new symptoms appear), accompanied by exorbitant amounts of testing. House's assistants provide the union of all high tech medicine & are capable of running any diagnostic under the sun (somehow, the hospital lacks lab techs!). By the end, the case is solved -- and more often than not the patient survives (a few lose the lottery).
One thing you actually DON'T see much of is DNA testing -- once in a while, but it hardly shows up as much as on a CSI/Law & Order type police procedural. DNA testing just doesn't televise well; the best you can do is show someone drawing their own blood (what, no buccal swabs?). In contrast, the MRI room has lots of fun angles -- private conversations behind the console, bouts of claustrophobia, or dramatic races to reach the suddenly stricken patient. Sequencers just aren't very dramatic.
So, I'm going to suggest an episode. Perhaps this qualifies as a "treatment" in Hollywood-speak. I have no desire for a career there, but if the writers take the idea I'd hardly turn down a walk-on.
A patient arrives at Princeton-Plainsboro seeking House due to a mysterious set of symptoms which has afflicted her for years. As usual with such, House is disdainful -- until the patient tries to hand him a DVD but dramatically collapses instead with some interesting symptom along the way. When the patient regains conciousness in a hospital bed, they start asking about the DVD again -- and then deliver the trump card: the DVD has her genome sequence on it.
House has no great interest in the DVD, and argues how useless it is. He's patently annoyed by it. One of the assistants makes the mistake of rising to the bait and proposing that perhaps a critical clue lies within -- and thereby gets assigned the task of cross-referencing EVERY polymorphism against the patient's symptoms. Several dead ends come from the DNA data, but nothing useful -- or in reality, just too many hypotheses which are too tenuous to do anything with. That doesn't stop the young assistants from batting some around and debating the now and future utility of such scans.
Now, as an aside, the story really (in my opinion) needs a complete genome scan. However, if there is a desire to garner some product placement that would narrow the candidates to one (Knome) at this stage. SNP scans are quite as dramatic!
At the end, the patient's puzzle is solved & they get to proceed in life knowing what they have & able to manage it. But, the kicker is that the assistant now cross-references the now known disease against the polymorphisms and comes up with an answer -- but it was buried deep within hundreds of other equally supported hypotheses. Finish the episode with some more back-and-forth amongst the characters about how this might play out the next time. How their careers might change. How well (or not so well) their training has prepared them for this.
House M.D. is an hourlong drama focusing on Dr. Gregory House, a brilliant diagnostician who is also an extremely difficult human being. He terrorizes his three junior colleagues, who are trapped in his orbit like the inner moons of Jupiter -- and subject to similar violent (though only psychologically) tidal forces. Three previous assistants have attained somewhat more distant orbits, though one has spiraled back in. His boss & a colleague attempt to be friends, but get much grief for their efforts.
As with most series TV, there is a basic formula, a framework which the writers decorate or modify each week, rarely breaking it entirely. The scheme here generally starts with a patient arriving with some strange, dramatic set of symptoms (usually exposited prior to the opening credits). House is either intrigued or blackmailed by his boss into taking the case Lots of diagnostic dead ends follow (and new symptoms appear), accompanied by exorbitant amounts of testing. House's assistants provide the union of all high tech medicine & are capable of running any diagnostic under the sun (somehow, the hospital lacks lab techs!). By the end, the case is solved -- and more often than not the patient survives (a few lose the lottery).
One thing you actually DON'T see much of is DNA testing -- once in a while, but it hardly shows up as much as on a CSI/Law & Order type police procedural. DNA testing just doesn't televise well; the best you can do is show someone drawing their own blood (what, no buccal swabs?). In contrast, the MRI room has lots of fun angles -- private conversations behind the console, bouts of claustrophobia, or dramatic races to reach the suddenly stricken patient. Sequencers just aren't very dramatic.
So, I'm going to suggest an episode. Perhaps this qualifies as a "treatment" in Hollywood-speak. I have no desire for a career there, but if the writers take the idea I'd hardly turn down a walk-on.
A patient arrives at Princeton-Plainsboro seeking House due to a mysterious set of symptoms which has afflicted her for years. As usual with such, House is disdainful -- until the patient tries to hand him a DVD but dramatically collapses instead with some interesting symptom along the way. When the patient regains conciousness in a hospital bed, they start asking about the DVD again -- and then deliver the trump card: the DVD has her genome sequence on it.
House has no great interest in the DVD, and argues how useless it is. He's patently annoyed by it. One of the assistants makes the mistake of rising to the bait and proposing that perhaps a critical clue lies within -- and thereby gets assigned the task of cross-referencing EVERY polymorphism against the patient's symptoms. Several dead ends come from the DNA data, but nothing useful -- or in reality, just too many hypotheses which are too tenuous to do anything with. That doesn't stop the young assistants from batting some around and debating the now and future utility of such scans.
Now, as an aside, the story really (in my opinion) needs a complete genome scan. However, if there is a desire to garner some product placement that would narrow the candidates to one (Knome) at this stage. SNP scans are quite as dramatic!
At the end, the patient's puzzle is solved & they get to proceed in life knowing what they have & able to manage it. But, the kicker is that the assistant now cross-references the now known disease against the polymorphisms and comes up with an answer -- but it was buried deep within hundreds of other equally supported hypotheses. Finish the episode with some more back-and-forth amongst the characters about how this might play out the next time. How their careers might change. How well (or not so well) their training has prepared them for this.
Subscribe to:
Posts (Atom)