My print version of Science showed up and the cover looks more like some foodie rag. I guessed wrong at first that they were peppers -- right family, wrong fruit. Nope, they are tomatoes.
I like growing tomatoes, though end-of-season output far outpaces my ability to consume them. It's fun growing different varieties, as there are so many different shapes, colors, sizes and flavors. Of course, all of the yard grown ones whip the store cardboard versions. On the other hand, it's hard to grow them around here this time of year -- though I once did have a rampant cherry tomato plant in the tearoom at Harvard (with bunsen burner supports & such staking it up!), though I didn't get any tomatoes until it dawned on me that I needed to play honeybee for the blossoms. When I interviewed at Millennium, someone had a tomato plant growing in the sequencing area (I'm sure the lab safety folks wouldn't let that happen any more!).
Anyway, the Science cover is for an interesting paper showing that a local gene duplication led to elongation of the fruit in one variety. Longer genes, longer fruit! The duplication is recent and was triggered by a retrotransposon, which altered the transcriptional environment around the gene. Cool!
Thursday, March 20, 2008
Thursday, February 21, 2008
History better learned late than never
I've always been interested in history, and the history of science is no exception. I thought I knew a bit about the history of DNA sequencing, so it was a bit of a rude surprise to read the obituaries on Wed for Dr. Ray Wu and discover that he had published one of the first DNA sequencing methods, a method that is credited with being the forerunner of Sanger sequencing. I was totally unaware of this history.
Sadly, the Wikipedia article on DNA sequencing doesn't cover this at all.
A bit of Medline hunting, aided by Dr. Wu's page at Cornell, found a few articles in PubMed, most sans abstracts and few with full text (Somebody PLEASE arrange legally to get classic J Mol Biol as free full text!). Luckily there are a few papers -- this NAR paper and an earlier PNAS one. If I'm reading it correctly, then it involved 2D analysis of digestion maps, which I had heard of so perhaps my historical knowledge isn't totally deficient.
The one question that occurs is why didn't Dr. Wu stay in the DNA sequencing business. I wonder if he left any thoughts -- was it just not interesting enough or did Sanger & Gilbert just jump ahead so he felt like it wasn't the right place to be. Whatever his reasons it can't really be criticized -- Wu had quite a publication record and appeared to be active essentially to the end of his life. It would just be interesting to understand why he took the direction he did.
Sadly, the Wikipedia article on DNA sequencing doesn't cover this at all.
A bit of Medline hunting, aided by Dr. Wu's page at Cornell, found a few articles in PubMed, most sans abstracts and few with full text (Somebody PLEASE arrange legally to get classic J Mol Biol as free full text!). Luckily there are a few papers -- this NAR paper and an earlier PNAS one. If I'm reading it correctly, then it involved 2D analysis of digestion maps, which I had heard of so perhaps my historical knowledge isn't totally deficient.
The one question that occurs is why didn't Dr. Wu stay in the DNA sequencing business. I wonder if he left any thoughts -- was it just not interesting enough or did Sanger & Gilbert just jump ahead so he felt like it wasn't the right place to be. Whatever his reasons it can't really be criticized -- Wu had quite a publication record and appeared to be active essentially to the end of his life. It would just be interesting to understand why he took the direction he did.
Friday, February 01, 2008
Microsoft's irritating limits
Okay, it’s really time to face facts – there’s no avoiding the issue. I am a creature of habit, but the good ones tend to wax and wane with a rarely changing backdrop of the bad ones. Posting regularly to this forum was a good habit, but one which has been scarcely seen for a while.
For the explanations, you can round up the usual suspects. Work pressures. Holiday madness. Routines once disturbed being hard to reform. But in the end, those are just excuses which must be actively combated. Or perhaps it is a lack of passion? And how can passion be rekindled? What will put a new burr in the saddle – well (to badly mix metaphors), perhaps something to get the blood boiling: a good old-fashioned screed.
Now, there are whole websites devoted to griping about Microsoft. Perhaps everything I say has been said before. And, perhaps it is all obsolete carping – for reasons good and perhaps not-so-good I’m still using the previous version of Office. But, given that most of these problems have persisted through so many versions during a decade-and-a-half, I’m not optimistic.
Now, I’ve picked a very broad target. Easy to hit. Like an archer trying to hit the ocean with an arrow while standing in the foam. But I’ll try to focus.
I could gripe almost endlessly about Bill’s Army’s graphical choices. How poor contrast, cluttered charts are the defaults. I won’t (today, at least) go into that, but suffice to say it would be unhealthy for everyone concerned if I were to be trapped in an elevator with the Excel Charts programming team.
No, today I won’t focus on cluttered vision but on a limited vision, when Microsoft has a good idea but then shackles it bizarrely. When it gives the user choices, but insists on giving them very few.
Excel is a workhorse for myself and my colleagues as a data organization, filtering & delivery tool. This has been a habit acquired way back in graduate school and carried through many iterations. I often gravitate towards Excel for taking structured notes, with it’s relentless row-and-column layout forcing useful order. It’s also easy to take those tables and move them into other programs, and in my current post I’ve discovered Excel’s facility for slurping data out of relational databases. All great stuff.
Now, once you’ve got a mess of data in Excel you’d like to sort, filter & highlight it. Naturally, Excel can sort data, the handy Auto-Filter can filter it, and conditional formatting can highlight cells. But they are all broken in the same way: for reasons numerological or otherwise you can do any of these on only 3 criteria each. Sort on three columns. Have three different states to filter on format on.
Now, to pretend to be fair, with Auto-Filter that restriction is per-column, so in a many columned spreadsheet I can filter each column with three criteria -- except when it gives me only two - -I swear it happens, though I can’t remember when , as I’m usually too busy swearing. But, if you wish to see in your dining spreadsheet when you ate Thai, Italian, German or Fusion, you are out-of-luck.
A work-around, of course, is to compute columns with groupings. This can be useful, especially since you can’t save those filter settings for reuse later, but such columns quickly become a visual headache. A few can be useful, but in my hypothetical dining spreadsheet I really won’t want one column for every possible combination of cuisines!
The visual issue brings up another half-way measure: pane splitting. A useful way to deal with so many columns or rows is to split the panes horizontally or vertically (or often both) – but you can have only one split! So if I want to compare data in columns A,Q and BC, I’m out of luck – unless I copy columns or laboriously hide (and later unhide) the columns in between. One help is to make the top row or first column so it doesn’t scroll offscreen – except that blocks out pane splitting!
Now, somebody might be churlish enough to suggest that these limits are rationally chosen and have to do with tradeoffs. Hogwash! Spotfire has a spectacular interface for limiting a field all different ways, with different idioms for choosing (checkboxes, sliders, radio buttons). And if you are having problems with splitting too many ways not working well with various windowing components (such as scrollbars), it’s not like the company doesn’t have its paws in both areas.
A related class of shortcomings, but not quite in the same category, comes around Excel’s functions. If you want to add or count things conditional on some column, there are the COUNTIF and SUMIF functions. But what if I want a conditional AVERAGE? Well, that at least can be gotten from COUNTIF and SUMIF – but no such luck if I want QUARTILE or MEDIAN or STDEV such ways. Why isn’t there a function APPLYIF?
Similarly, I’ve recently gotten hooked on Pivot Tables, which are handy for summarizing data. Basically, a Pivot Table can give you summary statistics on some cross-reference of fields in rows – for example, I could easily see in my eating out spreadsheet the sums of cuisines versus the months of the year. Naturally, these statistics include sum, count, average, stdev, min, mean – but no MEDIAN or QUARTILE . Aargh!
Alas, for the moment I must let my blood boil. When in a company environment, there is an advantage to having a lingua franca, even if it is a flawed one. Plus, what alternative do I have? Perhaps some of this is better in OpenOffice, but I’m not optimistic – my previous experience (though at least a year old) with OO was that it generally aped Microsoft’s flaws & introduced some new ones. Yeah, with more time and energy I could fix that myself, given that it’s open source, but wasn’t the limits of personal time and energy how I opened this article?
For the explanations, you can round up the usual suspects. Work pressures. Holiday madness. Routines once disturbed being hard to reform. But in the end, those are just excuses which must be actively combated. Or perhaps it is a lack of passion? And how can passion be rekindled? What will put a new burr in the saddle – well (to badly mix metaphors), perhaps something to get the blood boiling: a good old-fashioned screed.
Now, there are whole websites devoted to griping about Microsoft. Perhaps everything I say has been said before. And, perhaps it is all obsolete carping – for reasons good and perhaps not-so-good I’m still using the previous version of Office. But, given that most of these problems have persisted through so many versions during a decade-and-a-half, I’m not optimistic.
Now, I’ve picked a very broad target. Easy to hit. Like an archer trying to hit the ocean with an arrow while standing in the foam. But I’ll try to focus.
I could gripe almost endlessly about Bill’s Army’s graphical choices. How poor contrast, cluttered charts are the defaults. I won’t (today, at least) go into that, but suffice to say it would be unhealthy for everyone concerned if I were to be trapped in an elevator with the Excel Charts programming team.
No, today I won’t focus on cluttered vision but on a limited vision, when Microsoft has a good idea but then shackles it bizarrely. When it gives the user choices, but insists on giving them very few.
Excel is a workhorse for myself and my colleagues as a data organization, filtering & delivery tool. This has been a habit acquired way back in graduate school and carried through many iterations. I often gravitate towards Excel for taking structured notes, with it’s relentless row-and-column layout forcing useful order. It’s also easy to take those tables and move them into other programs, and in my current post I’ve discovered Excel’s facility for slurping data out of relational databases. All great stuff.
Now, once you’ve got a mess of data in Excel you’d like to sort, filter & highlight it. Naturally, Excel can sort data, the handy Auto-Filter can filter it, and conditional formatting can highlight cells. But they are all broken in the same way: for reasons numerological or otherwise you can do any of these on only 3 criteria each. Sort on three columns. Have three different states to filter on format on.
Now, to pretend to be fair, with Auto-Filter that restriction is per-column, so in a many columned spreadsheet I can filter each column with three criteria -- except when it gives me only two - -I swear it happens, though I can’t remember when , as I’m usually too busy swearing. But, if you wish to see in your dining spreadsheet when you ate Thai, Italian, German or Fusion, you are out-of-luck.
A work-around, of course, is to compute columns with groupings. This can be useful, especially since you can’t save those filter settings for reuse later, but such columns quickly become a visual headache. A few can be useful, but in my hypothetical dining spreadsheet I really won’t want one column for every possible combination of cuisines!
The visual issue brings up another half-way measure: pane splitting. A useful way to deal with so many columns or rows is to split the panes horizontally or vertically (or often both) – but you can have only one split! So if I want to compare data in columns A,Q and BC, I’m out of luck – unless I copy columns or laboriously hide (and later unhide) the columns in between. One help is to make the top row or first column so it doesn’t scroll offscreen – except that blocks out pane splitting!
Now, somebody might be churlish enough to suggest that these limits are rationally chosen and have to do with tradeoffs. Hogwash! Spotfire has a spectacular interface for limiting a field all different ways, with different idioms for choosing (checkboxes, sliders, radio buttons). And if you are having problems with splitting too many ways not working well with various windowing components (such as scrollbars), it’s not like the company doesn’t have its paws in both areas.
A related class of shortcomings, but not quite in the same category, comes around Excel’s functions. If you want to add or count things conditional on some column, there are the COUNTIF and SUMIF functions. But what if I want a conditional AVERAGE? Well, that at least can be gotten from COUNTIF and SUMIF – but no such luck if I want QUARTILE or MEDIAN or STDEV such ways. Why isn’t there a function APPLYIF?
Similarly, I’ve recently gotten hooked on Pivot Tables, which are handy for summarizing data. Basically, a Pivot Table can give you summary statistics on some cross-reference of fields in rows – for example, I could easily see in my eating out spreadsheet the sums of cuisines versus the months of the year. Naturally, these statistics include sum, count, average, stdev, min, mean – but no MEDIAN or QUARTILE . Aargh!
Alas, for the moment I must let my blood boil. When in a company environment, there is an advantage to having a lingua franca, even if it is a flawed one. Plus, what alternative do I have? Perhaps some of this is better in OpenOffice, but I’m not optimistic – my previous experience (though at least a year old) with OO was that it generally aped Microsoft’s flaws & introduced some new ones. Yeah, with more time and energy I could fix that myself, given that it’s open source, but wasn’t the limits of personal time and energy how I opened this article?
Thursday, January 17, 2008
Portrait of a scientist
Judah Folkman's passing this week was a sad event. I got to hear the great man speak -- at Millennium's spectacular internal seminar series 'Innovators in Medicine' -- and he was great -- inspiring, friendly, insightful. He will be missed.
Folkman was well known for having enduring brutal ridicule around his angiogenesis hypothesis. In his talk, he mentioned that he kept a room at his lab wallpapered with the years of rejection notices for grant applications and journal submission, using it to remind his students and staff that such difficulties could be overcome. One of the obits even published the crack that angiogenic factors exist 'only in the mind of the principal investigator'.
It was good to see the Globe reprint, although buried inside a section (the lead on the obituary did make the bottom of the front page) a wonderful portrait of Folkman. I don't know how long the article will remain free, but a smaller image shows up with the search. Folkman is shown framed by a lab bench with a face mixing weariness and thought. He did have quite a face! Not a movie star face, but one rich in features and details.
It isn't easy to think of good scientific portraits. Most departmental shots are straight formal shots, if not near mugshots. A few sterotyped poses predominate: for example, googling on Folkman yields some standards: hand on chin thoughtful, looking up from a microscope, teaching in front of a blackboard, listening to a colleague.
One classic is the shot of Watson & Crick with their model, though the photographer had Crick point with a slide rule, a tool utterly alien to the work. Any favorites out there? What are the best photos which illustrate the character and nature of a scientist?
Folkman was well known for having enduring brutal ridicule around his angiogenesis hypothesis. In his talk, he mentioned that he kept a room at his lab wallpapered with the years of rejection notices for grant applications and journal submission, using it to remind his students and staff that such difficulties could be overcome. One of the obits even published the crack that angiogenic factors exist 'only in the mind of the principal investigator'.
It was good to see the Globe reprint, although buried inside a section (the lead on the obituary did make the bottom of the front page) a wonderful portrait of Folkman. I don't know how long the article will remain free, but a smaller image shows up with the search. Folkman is shown framed by a lab bench with a face mixing weariness and thought. He did have quite a face! Not a movie star face, but one rich in features and details.
It isn't easy to think of good scientific portraits. Most departmental shots are straight formal shots, if not near mugshots. A few sterotyped poses predominate: for example, googling on Folkman yields some standards: hand on chin thoughtful, looking up from a microscope, teaching in front of a blackboard, listening to a colleague.
One classic is the shot of Watson & Crick with their model, though the photographer had Crick point with a slide rule, a tool utterly alien to the work. Any favorites out there? What are the best photos which illustrate the character and nature of a scientist?
Tuesday, January 15, 2008
Watching the returns
I'm probably going to stay up later than I should tonight watching the returns. Yep, I'm a junkie for the stuff -- it's just so exciting seeing history being made. The multitude have spoken, and now it is just time to watch.
However, I don't mean paying attention to my ex-governor and former first lady in a bipartite Midwestern state. I was a serious political junkie as a youth, but that has cooled -- not that it isn't important. No, these are the really important returns: new sequence data.
I've been involved with one project here for a long time, with a leading role. The project has been interesting -- as in the bittersweet toast "May you live in interesting times". Ups & downs, trials & tribulations, lights at the ends of tunnels that turned out to be oncoming freights. All sorts of traps I laid for myself & walked into.
But now, after all that, the thing is actually WORKING! Well, maybe. 'Cause the proof is in the pudding, and the pudding is made of BigDye. In Sanger we trust.
So, I have my little database query set up & every so often I ping for an update. Now, the great sequencing system runs autonomously & in batches, and I don't have the wherewithal from home to see where each of my precious plates is in the stacker (the only sure way is to examine the apparatus & we don't have a webcam installed), so an element of suspense is present. I know how many reads were called up for each clone, so I'll know when it's done.
So I sit here, trying to resist pinging again. It's unlikely that another plate has loaded & processed since the last ping. Just hold out one more minute. NOW! Nope, no new data. I can wait. Sure I can. Just another minute...
However, I don't mean paying attention to my ex-governor and former first lady in a bipartite Midwestern state. I was a serious political junkie as a youth, but that has cooled -- not that it isn't important. No, these are the really important returns: new sequence data.
I've been involved with one project here for a long time, with a leading role. The project has been interesting -- as in the bittersweet toast "May you live in interesting times". Ups & downs, trials & tribulations, lights at the ends of tunnels that turned out to be oncoming freights. All sorts of traps I laid for myself & walked into.
But now, after all that, the thing is actually WORKING! Well, maybe. 'Cause the proof is in the pudding, and the pudding is made of BigDye. In Sanger we trust.
So, I have my little database query set up & every so often I ping for an update. Now, the great sequencing system runs autonomously & in batches, and I don't have the wherewithal from home to see where each of my precious plates is in the stacker (the only sure way is to examine the apparatus & we don't have a webcam installed), so an element of suspense is present. I know how many reads were called up for each clone, so I'll know when it's done.
So I sit here, trying to resist pinging again. It's unlikely that another plate has loaded & processed since the last ping. Just hold out one more minute. NOW! Nope, no new data. I can wait. Sure I can. Just another minute...
Thursday, December 27, 2007
A Vertex by the sea?
If one thinks like a builder, it is not difficult to scan the prime biotech zone in Cambridge and see it full at some point. I remember bicycling to the Harvard Medical School in the 90's past an empty zone with just a couple of lone buildings; those buildings are now thickly surrounded, save some parkland. There are still some parking lots that might be made over, but in general there isn't a lot of free space left. Some single-story buildings might go (someone must be eyeing the boarded up saloon up the street from Novartis, but not those close to residential land -- which is a lot of them -- and there isn't much room up. Between a general Cantabrigian disdain for high rises & fire department restrictions on where labs can go, up is not a great option for biotech.
Throughout the zone there are also other uses for what space there is. MIT owns much of the land, and must be wondering whether it will be hemmed in. Urban planning has shifted away towards favoring a variety of uses, and so some of the new development in the zone has gone to residences, hotels & shops -- a good thing, too! Hopefully ways will be found to preserve some of the grittier older businesses, the car repair shops & such that are so convenient. But space must be found, or the biotech industry will stagnate.
There is a lot of open space to the far east, where once a large railroad yard sat in the netherlands between Cambridge and Charlestown. New buildings are springing up there & the developers have already advertised in biotech real estate sections.
However, others are thinking of a really big conceptual leap. In a Globe article before Christmas it was revealed that Vertex is contemplating moving their entire operation to new buildings to be constructed at Fan Pier. This is an area just off the center of Boston and on the waterfront, and which is in an area which is becoming a magnet for development. Better road connections, thanks to the Big Dig, a new transit line, a new federal courthouse, and the new convention center have led to other businesses, such as restaurants and hotels.
It's not hard to see the attraction of the place. Walkable to downtown Boston and a short walk to the transit hub (intercity & commuter train, bus, subway) at South Station. Within site (across the water, traversed by a tunnel) of the airport.
The obvious uses of this space were offices (particularly legal ones; the courthouse is next door) -- but biotech? I wouldn't have thought of it, but somebody did. It's a bold move, one to announce that Vertex has arrived as a FIPCO (Fully Integrated Pharmaceutical Company). The developer has already started acquiring permits around buildings suitable for lab space. And the location has other perks -- nearby Red Line access to the Harvard & MIT campuses, so it's almost like being in Cambridge. The new transit line doesn't yet go many places, but if a proposed tunnel is dug it could connect to the Longwood Hospitals area.
Despite all the hubbub in Cambridge, Boston itself doesn't host much biotech. I think there is some incubator-type space over in Charlestown and maybe some bits elsewhere, but mostly the main city plays a subsidiary role. Remote sites such as a derelict state hospital have sometimes been proposed, but nothing much has happened -- perhaps this could jump-start other unconventional locations for biotech in the Hub of the Universe.
Throughout the zone there are also other uses for what space there is. MIT owns much of the land, and must be wondering whether it will be hemmed in. Urban planning has shifted away towards favoring a variety of uses, and so some of the new development in the zone has gone to residences, hotels & shops -- a good thing, too! Hopefully ways will be found to preserve some of the grittier older businesses, the car repair shops & such that are so convenient. But space must be found, or the biotech industry will stagnate.
There is a lot of open space to the far east, where once a large railroad yard sat in the netherlands between Cambridge and Charlestown. New buildings are springing up there & the developers have already advertised in biotech real estate sections.
However, others are thinking of a really big conceptual leap. In a Globe article before Christmas it was revealed that Vertex is contemplating moving their entire operation to new buildings to be constructed at Fan Pier. This is an area just off the center of Boston and on the waterfront, and which is in an area which is becoming a magnet for development. Better road connections, thanks to the Big Dig, a new transit line, a new federal courthouse, and the new convention center have led to other businesses, such as restaurants and hotels.
It's not hard to see the attraction of the place. Walkable to downtown Boston and a short walk to the transit hub (intercity & commuter train, bus, subway) at South Station. Within site (across the water, traversed by a tunnel) of the airport.
The obvious uses of this space were offices (particularly legal ones; the courthouse is next door) -- but biotech? I wouldn't have thought of it, but somebody did. It's a bold move, one to announce that Vertex has arrived as a FIPCO (Fully Integrated Pharmaceutical Company). The developer has already started acquiring permits around buildings suitable for lab space. And the location has other perks -- nearby Red Line access to the Harvard & MIT campuses, so it's almost like being in Cambridge. The new transit line doesn't yet go many places, but if a proposed tunnel is dug it could connect to the Longwood Hospitals area.
Despite all the hubbub in Cambridge, Boston itself doesn't host much biotech. I think there is some incubator-type space over in Charlestown and maybe some bits elsewhere, but mostly the main city plays a subsidiary role. Remote sites such as a derelict state hospital have sometimes been proposed, but nothing much has happened -- perhaps this could jump-start other unconventional locations for biotech in the Hub of the Universe.
Thursday, December 20, 2007
Getting my hands wet
My undergraduate days were full of laboratory work. I got involved in student research as an undergraduate, spent a xmemorable summer being unsuccessful at DNA sequencing, and enrolled in a new very laboratory-intensive program. Between my mishaps in my coursework & my mishaps in my independent work, eventually it was suggested (and I didn't need much convincing) that maybe I could combine my hobby of computer programming with my focus on biology in some useful way.
After that, labs were on the radar intermittently. My department at Harvard insisted I do one 'wet' rotation, and so I spent a very enjoyable spring in a fly lab, mostly sorting flies & mutagenizing some but also mapping some transposon insertions by in situ hybridization. I also was a teaching assistant, and one year that meant running lab sections. My committee once seemed on the verge of insisting I do some lab work & I started cooking up a suitable experiment involving in vivo footprinting of DNA binding factor sequence preferences, but nothing ever happened.
At Millennium I failed to follow up on a golden opportunity, one that I rue to this day. Soon after joining one of the senior Mass Spec guys invited me to consider spending some time learning the equipment. Now, I had used a 'toy' mass spec as a high school intern to find the leak in some high vacuum equipment (successfully, which led to figuring out that you-know-who had inadvertantly put it there!) -- the spec is built into the instrument & you run a Pasteur pipet hooked to a helium tank around all the suspect spots -- when you see the helium spike, you've found the leak. Plus, my father worked on electronics for a mass spec that got a one-way trip into the Jovian atmosphere, so I had an interest in the things. Alas, I never quite followed through.
Later, it was suggested & I had started making concrete plans to learn how to work the HTS screening robotics at Millennium. Those plans were being laid when I got the pink slip (alas, they don't actually print the letter on pink paper!).
So, it is with some pride I can report I actually did some lab work yesterday. After several polite invitations from our headmistress of sequencing, I spent a bit of time observing & doing some 'scut work' -- having asked for as much as they dared give me. I labeled plates, recored them into the Laboratory Information Management System (LIMS) & later helped load & unload the robot preparing them. I sealed the plates (with a polite admonition that I was being too gentle with the roller). I also got to load the thermocyclers, select the right program & then unload them when done.
Not exactly the solo conquering of megabases that my imagination whipped up, but still very informative. You don't just pour DNA in one end of the sequencer and get data out the wires at the other end; there's a lot of manual tracking & care involved, even in a highly automated modern sequencing laboratory. It actually turns out that the process I observed will be upended in the near future for a much less hands-on, much higher throughput one. And, I couldn't help thinking of the various next-gen technologies, where you just do away with all those 96 or 384 well plates & instead run much larger numbers of sequencing reactions in a much smaller area.
It's all very different than when I was trying to use Sequenase, single channel pipettors, radioactive S35 & slab gels nineteen years ago. And one difference in particular relates to the title of this entry -- no water baths! You don't actually get your hands wet anymore -- nice dry thermocyclers do all the incubating instead.
After that, labs were on the radar intermittently. My department at Harvard insisted I do one 'wet' rotation, and so I spent a very enjoyable spring in a fly lab, mostly sorting flies & mutagenizing some but also mapping some transposon insertions by in situ hybridization. I also was a teaching assistant, and one year that meant running lab sections. My committee once seemed on the verge of insisting I do some lab work & I started cooking up a suitable experiment involving in vivo footprinting of DNA binding factor sequence preferences, but nothing ever happened.
At Millennium I failed to follow up on a golden opportunity, one that I rue to this day. Soon after joining one of the senior Mass Spec guys invited me to consider spending some time learning the equipment. Now, I had used a 'toy' mass spec as a high school intern to find the leak in some high vacuum equipment (successfully, which led to figuring out that you-know-who had inadvertantly put it there!) -- the spec is built into the instrument & you run a Pasteur pipet hooked to a helium tank around all the suspect spots -- when you see the helium spike, you've found the leak. Plus, my father worked on electronics for a mass spec that got a one-way trip into the Jovian atmosphere, so I had an interest in the things. Alas, I never quite followed through.
Later, it was suggested & I had started making concrete plans to learn how to work the HTS screening robotics at Millennium. Those plans were being laid when I got the pink slip (alas, they don't actually print the letter on pink paper!).
So, it is with some pride I can report I actually did some lab work yesterday. After several polite invitations from our headmistress of sequencing, I spent a bit of time observing & doing some 'scut work' -- having asked for as much as they dared give me. I labeled plates, recored them into the Laboratory Information Management System (LIMS) & later helped load & unload the robot preparing them. I sealed the plates (with a polite admonition that I was being too gentle with the roller). I also got to load the thermocyclers, select the right program & then unload them when done.
Not exactly the solo conquering of megabases that my imagination whipped up, but still very informative. You don't just pour DNA in one end of the sequencer and get data out the wires at the other end; there's a lot of manual tracking & care involved, even in a highly automated modern sequencing laboratory. It actually turns out that the process I observed will be upended in the near future for a much less hands-on, much higher throughput one. And, I couldn't help thinking of the various next-gen technologies, where you just do away with all those 96 or 384 well plates & instead run much larger numbers of sequencing reactions in a much smaller area.
It's all very different than when I was trying to use Sequenase, single channel pipettors, radioactive S35 & slab gels nineteen years ago. And one difference in particular relates to the title of this entry -- no water baths! You don't actually get your hands wet anymore -- nice dry thermocyclers do all the incubating instead.
Monday, December 17, 2007
The Nine Circles of Basel
Human society has long had a fascination with large human enterprises. Complex governmental systems, with ministers and sub-ministers, seem to appear in the literature of each civilization (at least the few I have sampled). Confucius's Analects are largely about how to properly govern. Those nine rings of Dante's Inferno don't just run themselves, but clearly have a hierarchy of devils managing them. Throughout history, churches, governments and the military have had organized systems of managers and sub-managers. While these are often the bane of our existence, we do occasionally get such joys as the novel Catch-22 or the movie Brazil -- and the terrors of 1984. (Does some professor somewhere teach a course in the Literature of Bureaucracy?)
According to the few sources I have read, modern corporate bureaucracy can largely trace its origins to the boom of railroads in the mid-1800's. Railroads required organization, or at a minimum freight would rot in the wrong place and at worst expensive rolling stock would be destroyed in collisions (the loss of the passengers & crew being not much concern of the railroad barons). Many of the American railroads were run, particularly after the American Civil War, by men with strong military experience, and so organized structures grew. And grew. And grew.
The Wall Street Journal carries an item (which, as the one silver lining to the Murdoch takeover, is free) that a key focus of Novartis' announced job cuts is to eliminate bureaucracy. Dr. Vasella, the CEO, was shocked (shocked!) that a mid-level manager in one Novartis division had 6 levels of employees below him, and believes this is too many.
I used to joke at Millennium that I was routinely being demoted. This was obviously so, as the number of people between myself and Mark Levin kept growing. When I joined, around 250 employees, there were two people interposed, but at times it was at least four or five. I reported to a group leader, who reported to the informatics head, who reported to the technology head, who reported to the CSO, who reported to Mark. At various times the higher levels would shift, but like the beach which does the same it rarely changed my routine existence. We'd get invited to different meetings, my impertinent emails would irritate different superiors, but overall it took some digging to find the real changes -- and by the time you did, the next reorganization would be upon us.
Six levels at first glance seems like a lot. If each manager had 10 reports, then that's a million employees, right? Much more than the employment of Novartis (calculated from another report at around 100K. Vasella has apparently decreed that no division shall have more than six layers of reporting (hmm, with a few layers at HQ, that would be how many levels total? :-). Suffering from the sclerotic plaques of bureaucracy? Apply the statin of restructuring!
I'm no fan of bureaucracy -- I have a particular talent for botching official forms -- but I hope that Vasella & his staff think carefully about the unintended side-effects of such a crusade. Too many layers stifle innovation. But too few may have consequences as well.
My jaundiced history of corporate organization leaves out some of the other drivers of layers. Yes, 6 layers should be sufficient to support 1 million employees -- if each manager has precisely 10 reports. Even that many is too many for some Human Resource experts' tastes, but more importantly sometimes a manager should have fewer. Someone might be a great manager of three but horrible at seven. Or you do really need 3 mass spectroscopists managed by one senior one -- but that's it. Also, some layers of management are a way to train & retain valued employees.
At first thought, the main danger of a blanket edict is that the organization will adjusted to fit the management dictum, not the other way round. Procrustes as organizational expert. While that is certainly not a foreign mode of operation for large companies, it is hardly what you want to encourage! Second, the organization will be stifled in new ways: sorry, we can't enlarge this successful organization without blowing it up, as we've hit our depth chart limit. Furthermore, the asterisks are likely to start rolling out -- and with them additional warpings of the original goal. Summer interns -- they don't count. Full-time contractors -- nah. Outsourcing -- well, of course not! And as routes to avoid limits are found, they will be used -- whether they make overall sense or not.
Is there an alternative? It's hard to say. Like most management mantras this paring, if applied judiciously, might be a good thing. Loosening up rules to minimize how far purchasing requests must percolate upwards are good. Identifying where additional layers are not adding expertise but only inertia is good -- but inertia tends to be in the eye of the beholder (or perhaps, bestopper). Legal requires such as Sarbane-Oxley don't exactly encourage a free hand either -- the shareholders like to know how you're spending their money, and if not them then Capitol Hill.
Of course, one solution is to work in a very small company. Then there can't be too many layers between you and the top, unless the company has a completely linear organizational structure! That's not to say that small companies don't face the same human & informational challenges or solve them easily, but too many layers of managers tends to be low on the worry list.
According to the few sources I have read, modern corporate bureaucracy can largely trace its origins to the boom of railroads in the mid-1800's. Railroads required organization, or at a minimum freight would rot in the wrong place and at worst expensive rolling stock would be destroyed in collisions (the loss of the passengers & crew being not much concern of the railroad barons). Many of the American railroads were run, particularly after the American Civil War, by men with strong military experience, and so organized structures grew. And grew. And grew.
The Wall Street Journal carries an item (which, as the one silver lining to the Murdoch takeover, is free) that a key focus of Novartis' announced job cuts is to eliminate bureaucracy. Dr. Vasella, the CEO, was shocked (shocked!) that a mid-level manager in one Novartis division had 6 levels of employees below him, and believes this is too many.
I used to joke at Millennium that I was routinely being demoted. This was obviously so, as the number of people between myself and Mark Levin kept growing. When I joined, around 250 employees, there were two people interposed, but at times it was at least four or five. I reported to a group leader, who reported to the informatics head, who reported to the technology head, who reported to the CSO, who reported to Mark. At various times the higher levels would shift, but like the beach which does the same it rarely changed my routine existence. We'd get invited to different meetings, my impertinent emails would irritate different superiors, but overall it took some digging to find the real changes -- and by the time you did, the next reorganization would be upon us.
Six levels at first glance seems like a lot. If each manager had 10 reports, then that's a million employees, right? Much more than the employment of Novartis (calculated from another report at around 100K. Vasella has apparently decreed that no division shall have more than six layers of reporting (hmm, with a few layers at HQ, that would be how many levels total? :-). Suffering from the sclerotic plaques of bureaucracy? Apply the statin of restructuring!
I'm no fan of bureaucracy -- I have a particular talent for botching official forms -- but I hope that Vasella & his staff think carefully about the unintended side-effects of such a crusade. Too many layers stifle innovation. But too few may have consequences as well.
My jaundiced history of corporate organization leaves out some of the other drivers of layers. Yes, 6 layers should be sufficient to support 1 million employees -- if each manager has precisely 10 reports. Even that many is too many for some Human Resource experts' tastes, but more importantly sometimes a manager should have fewer. Someone might be a great manager of three but horrible at seven. Or you do really need 3 mass spectroscopists managed by one senior one -- but that's it. Also, some layers of management are a way to train & retain valued employees.
At first thought, the main danger of a blanket edict is that the organization will adjusted to fit the management dictum, not the other way round. Procrustes as organizational expert. While that is certainly not a foreign mode of operation for large companies, it is hardly what you want to encourage! Second, the organization will be stifled in new ways: sorry, we can't enlarge this successful organization without blowing it up, as we've hit our depth chart limit. Furthermore, the asterisks are likely to start rolling out -- and with them additional warpings of the original goal. Summer interns -- they don't count. Full-time contractors -- nah. Outsourcing -- well, of course not! And as routes to avoid limits are found, they will be used -- whether they make overall sense or not.
Is there an alternative? It's hard to say. Like most management mantras this paring, if applied judiciously, might be a good thing. Loosening up rules to minimize how far purchasing requests must percolate upwards are good. Identifying where additional layers are not adding expertise but only inertia is good -- but inertia tends to be in the eye of the beholder (or perhaps, bestopper). Legal requires such as Sarbane-Oxley don't exactly encourage a free hand either -- the shareholders like to know how you're spending their money, and if not them then Capitol Hill.
Of course, one solution is to work in a very small company. Then there can't be too many layers between you and the top, unless the company has a completely linear organizational structure! That's not to say that small companies don't face the same human & informational challenges or solve them easily, but too many layers of managers tends to be low on the worry list.
Thursday, December 13, 2007
Corporate DNA
Hsien-Hsien is frequently asking 'What's in your DNA?', and a correspondent of mine recently pointed out that "it's in our DNA" is becoming a bit of a corporate cliche. Indeed, there is at least one TV advertising campaign on those lines, though the ad writers would be disappointed to find out I can't name the company.
Now, I'm not always responsive to suggestions for blog posts, but since the writer shares both my mitochondrial & Y-chromosome genotypes, I was more easily swayed.
The question posed is this: what do companies asking this really mean, or more specifically what might it mean that they don't intend (very Dilbert-esque). Presumably they are trying to make a statement about deeply embedded values, but what does it really mean to have something in your DNA? For example, do they mean to imply:
Now, many of these statements may well be true about a given company, but is that what you really want to project?
What's in my company's DNA? Well, that's easy -- it's what the customer ordered! :-)
Now, I'm not always responsive to suggestions for blog posts, but since the writer shares both my mitochondrial & Y-chromosome genotypes, I was more easily swayed.
The question posed is this: what do companies asking this really mean, or more specifically what might it mean that they don't intend (very Dilbert-esque). Presumably they are trying to make a statement about deeply embedded values, but what does it really mean to have something in your DNA? For example, do they mean to imply:
- A lot of our company is unfathomable to the human mind
- There's a lot of redundancy here
- Often we often repeat ourselves often repeatedly, often repeating repetitiously.
- We retain bits of those who invade our corporate DNA, though with not much rhyme or reason
- A lot of pieces of the organization resemble decayed portions of other pieces of our organization
- Some pieces of our organization are non-functional, though they closely resemble functional pieces of related organizations
- Most of our organization has no immediate impact on routine operations, or emergency ones
- Most of our organization has no immediate obvious purpose, if any
- Our corporate practices are not the best designable, but rather reflect an accumulation of historical accidents
Now, many of these statements may well be true about a given company, but is that what you really want to project?
What's in my company's DNA? Well, that's easy -- it's what the customer ordered! :-)
Monday, December 03, 2007
What Cell Is This, Who Laid To Rest?
With the holiday season upon us, I break out my seasonal music. So it is only fitting while I ponder various versions of Greensleeves that the NIH is finally starting to lay down the lay about the identity of cell cultures.
Cell cultures are great, but the problem is most of the cells pretty much look the same -- especially when growing in a culture dish. Now, I'm sure folks expert in the field will say they can distinguish many of them even without a microscope, but the fact remains that errors have frequently occurred and more than a few published studies are wrong because their cell culture wasn't what they thought it was. At my previous posting we burned a lot of effort on one large dataset that turned out to be useless for just this reason.
The prod in this case was an open letter by a number of researchers, and the response will apparently be to encourage referees to downgrade grant proposals and such which do not authenticate their cultures. 'bout time.
At Millennium we had licensed the Ingenuity database, which is a spectacular collection of biomedical facts culled from the primary literature (spectacular, but neither perfectly correct nor comprehensive, but amazing nonetheless). When gearing up for a big experiment on cell line X, we might try to pull all the knowledge of the database derived from experiments on cell line X -- the level of detail which Ingenuity provides. Of course, some of these would be contradictory -- and I found other cases where two published experiments were claimed to be precisely the same, but with very different results. The letter cites estimates that 20% of cell cultures are the wrong thing, which might explain a few of these.
Particularly in cancer research, this lack of a critical control is downright stupid. Nowadays, a somewhat pricey but powerful cell typing method are SNP chips. These will clearly disambiguate mouse from human (you won't get much signal if from the wrong species!), but can also probe deletions & copy number variation (though not balanced translocations). Given that most tumor cell lines are pretty fouled up in the DNA operations & maintenance department, expecting these cell lines to be stable is pretty unreasonable. Particularly if the experimenter is deliberately selecting for something (such as adherent growth, or growth in the absence of specific factors, or such), checking for major changes just makes sense -- and will also catch sneaky invaders who might take over such a culture.
Science published a good news article on the topic last winter -- alas, it will require subscription access.
Cell cultures are great, but the problem is most of the cells pretty much look the same -- especially when growing in a culture dish. Now, I'm sure folks expert in the field will say they can distinguish many of them even without a microscope, but the fact remains that errors have frequently occurred and more than a few published studies are wrong because their cell culture wasn't what they thought it was. At my previous posting we burned a lot of effort on one large dataset that turned out to be useless for just this reason.
The prod in this case was an open letter by a number of researchers, and the response will apparently be to encourage referees to downgrade grant proposals and such which do not authenticate their cultures. 'bout time.
At Millennium we had licensed the Ingenuity database, which is a spectacular collection of biomedical facts culled from the primary literature (spectacular, but neither perfectly correct nor comprehensive, but amazing nonetheless). When gearing up for a big experiment on cell line X, we might try to pull all the knowledge of the database derived from experiments on cell line X -- the level of detail which Ingenuity provides. Of course, some of these would be contradictory -- and I found other cases where two published experiments were claimed to be precisely the same, but with very different results. The letter cites estimates that 20% of cell cultures are the wrong thing, which might explain a few of these.
Particularly in cancer research, this lack of a critical control is downright stupid. Nowadays, a somewhat pricey but powerful cell typing method are SNP chips. These will clearly disambiguate mouse from human (you won't get much signal if from the wrong species!), but can also probe deletions & copy number variation (though not balanced translocations). Given that most tumor cell lines are pretty fouled up in the DNA operations & maintenance department, expecting these cell lines to be stable is pretty unreasonable. Particularly if the experimenter is deliberately selecting for something (such as adherent growth, or growth in the absence of specific factors, or such), checking for major changes just makes sense -- and will also catch sneaky invaders who might take over such a culture.
Science published a good news article on the topic last winter -- alas, it will require subscription access.
Sunday, December 02, 2007
Metastasis research deficit?
The Boston Globe had an article starting on the front page (briefly free, then will require $$) titled 'Critics blast slow progress on cancer: Say costly drugs do little to extend lives'. As with most newspaper articles the piece is short on hard data and longer on quotes. The most data-rich element is a graphic on the front page comparing 5-year survival rates for isolated vs. metastatic cancers (colon 89.8% vs. 10.3%; prostate: 100% vs. 11.9%; lung: 49.1% vs 3.0% & breast: 98.0% vs. 26.7%).
Two data items of interest that are here. First, an author who is "writing a book on the 'dysfunctional' cancer research industry' claims that 0.5% of federal research dollars have gone to studies of metastasis. The other is that 92% of cancer drugs entering human testing fail to make it to market.
If that 0.5% number is correct, it is most unfortunate. Early detection is great, but there needs to more focus on preventing metastases and treating them once they occur. There is a quote from Judah Folkman that some promising initial results have been seen using angiogenesis inhibitors to prevent metastasis, but they are clearly very small studies.
A key point that the article hammers on is that cancer researchers have constantly been promising that cures were around the corner, and yet that hasn't been realized. It cites the 36-year war on cancer, which was promoted as ending cancer by the bicentennial in 1976. More recently, now FDA commissioner Andrew von Eschenbach was proposing eradicating cancer by 2015.
What the article fails to explore is that we can't really know if we are about to have a sharp turnaround. As the article states, the long term survival mark is 5 years -- which means we won't know if the new drugs of 2007 had a real impact until pretty close to 2012. It also fails to explore the idea that extending lives by 'a few months' may be the initial signal, but that further optimization may extend that -- or that drugs are initially tested in desparately ill populations, where the deck may be highly stacked in the tumor's favor. In earlier patient populations, more notable gains may be practical.
Personally, I think that proposing to eradicate cancer by some time in the very near future is a recipe for disaster: again, given that 5-year survival is the key benchmark, eradicating cancer by 2015 would either mean (a) nearly perfect early detection [for cancers where that is nearly synonymous with a cure] and/or (b) eradicating cancer with the drugs in current late-stage testing, since only they could hit the clinics in a big way by 2010 so that 5-year survival could be measured by 2015. The former is just not realistic, and there's no great buzz from the industry that something is there to fulfill role b. Instead, researchers should set more reasonable expectations based on what is realistic. New tools for exploring cancer genomes & personalizing treatment will (IMHO) start making an impact -- but not for 5-10 years as they are tuned & troubleshot.
One other interesting note: amidst bemoaning flat U.S. government support for cancer research, it is noted that various patient-driven organizations are pumping money in or setting up key resources (such as a myeloma tissue bank set up by the Multiple Myeloma Research Foundation). What is interesting, and will hopefully continue, is some of these private organizations trying to invest in proposals that are kind of 'out there'. According to the article, the Komen foundation (breast cancer) has announced plans to invest $600M "to find wild ideas that will break the mold". That's some serious money, and if it is used to fund the 'unfundable' it will probably mean a lot of money going for failures -- and a few spectacular advances. If patients are impatient with research progress, funding what the establishment doesn't is a good way to express that frustration -- and maybe make a huge difference.
Two data items of interest that are here. First, an author who is "writing a book on the 'dysfunctional' cancer research industry' claims that 0.5% of federal research dollars have gone to studies of metastasis. The other is that 92% of cancer drugs entering human testing fail to make it to market.
If that 0.5% number is correct, it is most unfortunate. Early detection is great, but there needs to more focus on preventing metastases and treating them once they occur. There is a quote from Judah Folkman that some promising initial results have been seen using angiogenesis inhibitors to prevent metastasis, but they are clearly very small studies.
A key point that the article hammers on is that cancer researchers have constantly been promising that cures were around the corner, and yet that hasn't been realized. It cites the 36-year war on cancer, which was promoted as ending cancer by the bicentennial in 1976. More recently, now FDA commissioner Andrew von Eschenbach was proposing eradicating cancer by 2015.
What the article fails to explore is that we can't really know if we are about to have a sharp turnaround. As the article states, the long term survival mark is 5 years -- which means we won't know if the new drugs of 2007 had a real impact until pretty close to 2012. It also fails to explore the idea that extending lives by 'a few months' may be the initial signal, but that further optimization may extend that -- or that drugs are initially tested in desparately ill populations, where the deck may be highly stacked in the tumor's favor. In earlier patient populations, more notable gains may be practical.
Personally, I think that proposing to eradicate cancer by some time in the very near future is a recipe for disaster: again, given that 5-year survival is the key benchmark, eradicating cancer by 2015 would either mean (a) nearly perfect early detection [for cancers where that is nearly synonymous with a cure] and/or (b) eradicating cancer with the drugs in current late-stage testing, since only they could hit the clinics in a big way by 2010 so that 5-year survival could be measured by 2015. The former is just not realistic, and there's no great buzz from the industry that something is there to fulfill role b. Instead, researchers should set more reasonable expectations based on what is realistic. New tools for exploring cancer genomes & personalizing treatment will (IMHO) start making an impact -- but not for 5-10 years as they are tuned & troubleshot.
One other interesting note: amidst bemoaning flat U.S. government support for cancer research, it is noted that various patient-driven organizations are pumping money in or setting up key resources (such as a myeloma tissue bank set up by the Multiple Myeloma Research Foundation). What is interesting, and will hopefully continue, is some of these private organizations trying to invest in proposals that are kind of 'out there'. According to the article, the Komen foundation (breast cancer) has announced plans to invest $600M "to find wild ideas that will break the mold". That's some serious money, and if it is used to fund the 'unfundable' it will probably mean a lot of money going for failures -- and a few spectacular advances. If patients are impatient with research progress, funding what the establishment doesn't is a good way to express that frustration -- and maybe make a huge difference.
Wednesday, November 28, 2007
The Incredible Shrinking Human Genome
When the human genome was still terra incognito (or, at least our knowledge of the sequence was something like my view of the world sans my glasses oft mistaken for bulletproof glass), a key question was how many genes were present. It was widely cited by textbooks that the number was somewhere in the 50K-70K range, or perhaps even 100K, and some of the gene database companies such as Incyte and HGS and Hyseq were gleefully proclaiming the number much higher (just think what you are missing without our product!). The number wasn't unimportant. If you had some other estimate of what fraction of genes might be good targets for drug development, then the total number of drug targets was dependent on your estimate of the number of genes -- and drug targets were saleable -- and patentable.
At some point, a clever chap at Millennium decided to try to pin down these estimates. First he went for the textbook numbers, which everyone thought were well reasoned from old DNA melting curve experiments estimating the amount of non-repetitive DNA. Surprisingly, he was unable to find any solid calculation converting one to the other -- for all his searching, it appeared that the human gene estimate had appeared spontaneously like a quantum particle.
Using some other lines of thinking (I actually have a copy of his neat document somewhere, though technically it is a Millennium secret -- nothing just ages out of confidentiality. Silly, isn't that!) he argued from estimates of the gene content of yeast and from what had been found from C.elegans for a new estimate. Now, I couldn't find the flaw in his logic but I couldn't quite get myself to accept the estimate. It was preposterous! Only 30K genes for human?
Well, of course the estimate came in even further south of there. And a new paper from the Broad has nearly nipped that down to 20K even. Alas, the spectacularly endowed Broad wasn't munificent enough to publish with the Open Access option for PNAS, so until I make another pilgrimage to the MIT Library I'm stuck skimming the abstract, supporting materials & GenomeWeb writeup.
In some sense, the analysis is inevitable. It's hard to look at one genome and get an accurate gene estimate, but with so many mammalian genomes it gets easier -- and this paper apparently focused on primate genomes, which we have an amazing number of already. It sounds like they focused on ORFs found in human mRNA data, which at least removes the exon prediction problem.
The paper has the usual caveats. The genome is finished -- but not so finished. Bits and pieces are still getting polished up, and while they are generally dull and monotonous a gene or two might still hide there (the GenomeWeb bit mentions 197 genes found since the 'completion' of the genome which were omitted). The definition of gene is always tricky, generally going along the lines of Humpty Dumpty in Looking Glass: 'When I use a word...it means just what I choose it to mean -- neither more nor less.'. Gene here means protein-coding gene, to the exclusion of the RNA-only genes of seemingly endless flavor that pepper the genome.
The other class of caveat is very short ORFs -- and some very short ORFs do interesting things. For example, many neurotransmitters are synthesized from short ORFs -- and tend to evolve quickly, making it challenging to find them (I know, I tried in my past life).
Will this gene accounting ever end? The number will probably keep twiddling back and forth, but not by huge leaps barring some entirely new class of translational mechanism.
Speaking of genes & accounting, one of the little gags in Mr. Magorium's Wonder Emporium, a bit of movie fluff that is neither harmful nor wonderful, is a word derivation. The title character hires an accountant to assay his monetary worth, and promptly dissects the title: clearly it is a counting mutant. I find mutants more interesting than accountants, but both have their place -- and I never before realized that one was a subset of the other!
At some point, a clever chap at Millennium decided to try to pin down these estimates. First he went for the textbook numbers, which everyone thought were well reasoned from old DNA melting curve experiments estimating the amount of non-repetitive DNA. Surprisingly, he was unable to find any solid calculation converting one to the other -- for all his searching, it appeared that the human gene estimate had appeared spontaneously like a quantum particle.
Using some other lines of thinking (I actually have a copy of his neat document somewhere, though technically it is a Millennium secret -- nothing just ages out of confidentiality. Silly, isn't that!) he argued from estimates of the gene content of yeast and from what had been found from C.elegans for a new estimate. Now, I couldn't find the flaw in his logic but I couldn't quite get myself to accept the estimate. It was preposterous! Only 30K genes for human?
Well, of course the estimate came in even further south of there. And a new paper from the Broad has nearly nipped that down to 20K even. Alas, the spectacularly endowed Broad wasn't munificent enough to publish with the Open Access option for PNAS, so until I make another pilgrimage to the MIT Library I'm stuck skimming the abstract, supporting materials & GenomeWeb writeup.
In some sense, the analysis is inevitable. It's hard to look at one genome and get an accurate gene estimate, but with so many mammalian genomes it gets easier -- and this paper apparently focused on primate genomes, which we have an amazing number of already. It sounds like they focused on ORFs found in human mRNA data, which at least removes the exon prediction problem.
The paper has the usual caveats. The genome is finished -- but not so finished. Bits and pieces are still getting polished up, and while they are generally dull and monotonous a gene or two might still hide there (the GenomeWeb bit mentions 197 genes found since the 'completion' of the genome which were omitted). The definition of gene is always tricky, generally going along the lines of Humpty Dumpty in Looking Glass: 'When I use a word...it means just what I choose it to mean -- neither more nor less.'. Gene here means protein-coding gene, to the exclusion of the RNA-only genes of seemingly endless flavor that pepper the genome.
The other class of caveat is very short ORFs -- and some very short ORFs do interesting things. For example, many neurotransmitters are synthesized from short ORFs -- and tend to evolve quickly, making it challenging to find them (I know, I tried in my past life).
Will this gene accounting ever end? The number will probably keep twiddling back and forth, but not by huge leaps barring some entirely new class of translational mechanism.
Speaking of genes & accounting, one of the little gags in Mr. Magorium's Wonder Emporium, a bit of movie fluff that is neither harmful nor wonderful, is a word derivation. The title character hires an accountant to assay his monetary worth, and promptly dissects the title: clearly it is a counting mutant. I find mutants more interesting than accountants, but both have their place -- and I never before realized that one was a subset of the other!
Tuesday, November 20, 2007
Gene Logic successfully repositions, Ore What?
Gene Logic today announced that Pfizer has filed a patent based on a Gene Logic drug repositioning effort. This would appear to be one of the most significant votes of confidence in such efforts by an outside partner.
Drug repositioning is the idea of finding new therapeutic uses for advanced compounds, particularly compounds which are very advanced but failed due to poor efficacy in the originally targeted disease. A number of companies have sprung up in this field -- the two I am most familiar with are Gene Logic and Genstruct -- and at least some large pharmas have in-house programs.
The reality is that many existing drugs have origins in therapeutic areas which are quite different than those they started in. Perhaps the most notorious case is Viagra, which was muddling along as an anti-hypertensive until an unusual side effect was spotted. Minoxidil similarly began in the anti-hypertensive until its side effect was noted. The route to some psychiatric medications began with anti-tuberculosis agents and antihistamines. I doubt that's a complete list.
Gene Logic is one of the original cohort of genomics companies and has been through many iterations of business plan. If memory serves, they were one of several companies originally built around a differential display technology, a way of obtaining mRNA signatures for diseases which predated microarrays. Gene Logic later became one of the major players in the toxicogenomics space, and as part of that effort built a large in-house Affy-based microarray effort. They built microarray databases for a number of disease areas (I've used their oncology database), built a sizable bioinformatics effort, and even acquired their own CRO.
However, none of that could quite be converted into a stream of gold, so over the last year or so the whole mess has been deconstructed, leaving behind the drug repositioning business which had begun as a unit of Millennium (which is one reason I'm familiar with it). They'll even be changing their name soon, to Ore Pharmaceuticals (presumably Overburden and Slag, while appropriate for the mining theme, did not last long in the naming queue).
While there is certainly historical precedent for repositioning, the question remains whether companies can make money doing it, and whether those companies will simply be the big pharmas or the gaggle of biotechs chasing after the concept. Depending on the company, some mixture of in vivo models, in vitro models and computational methods are used. One way to think of it is doing drug discovery, but with a compound which already has safety data on it. There is also extensive interest in the concept in the academic sector, which is a very good thing -- many drugs which may be repositionable have little or no patent life yet, meaning companies will find it difficult to invest in them with any hope for a return.
Gene Logic / Ore has one repositioned drug which has gone through clinical trials, GL1001 (nee MLN4760). This is a drug originally developed by Millennium as an ACE2 inhibitor. Since I'm among the discoverers of ACE2, I tend to key an eye on this one. Millennium gave it a whirl in obesity, but now Gene Logic has found a signal in inflammatory bowel disease in animal models.
That Pfizer bothered to file a patent is significant, as it triggered a milestone payment -- amount unspecified, but these are usually something interesting. But that is still a long way from starting a new trial -- that will be the real milestone, and whichever drug repositioning firm can claim that will really be crowing -- that is, until somebody actually gets a drug approved this way.
Drug repositioning is the idea of finding new therapeutic uses for advanced compounds, particularly compounds which are very advanced but failed due to poor efficacy in the originally targeted disease. A number of companies have sprung up in this field -- the two I am most familiar with are Gene Logic and Genstruct -- and at least some large pharmas have in-house programs.
The reality is that many existing drugs have origins in therapeutic areas which are quite different than those they started in. Perhaps the most notorious case is Viagra, which was muddling along as an anti-hypertensive until an unusual side effect was spotted. Minoxidil similarly began in the anti-hypertensive until its side effect was noted. The route to some psychiatric medications began with anti-tuberculosis agents and antihistamines. I doubt that's a complete list.
Gene Logic is one of the original cohort of genomics companies and has been through many iterations of business plan. If memory serves, they were one of several companies originally built around a differential display technology, a way of obtaining mRNA signatures for diseases which predated microarrays. Gene Logic later became one of the major players in the toxicogenomics space, and as part of that effort built a large in-house Affy-based microarray effort. They built microarray databases for a number of disease areas (I've used their oncology database), built a sizable bioinformatics effort, and even acquired their own CRO.
However, none of that could quite be converted into a stream of gold, so over the last year or so the whole mess has been deconstructed, leaving behind the drug repositioning business which had begun as a unit of Millennium (which is one reason I'm familiar with it). They'll even be changing their name soon, to Ore Pharmaceuticals (presumably Overburden and Slag, while appropriate for the mining theme, did not last long in the naming queue).
While there is certainly historical precedent for repositioning, the question remains whether companies can make money doing it, and whether those companies will simply be the big pharmas or the gaggle of biotechs chasing after the concept. Depending on the company, some mixture of in vivo models, in vitro models and computational methods are used. One way to think of it is doing drug discovery, but with a compound which already has safety data on it. There is also extensive interest in the concept in the academic sector, which is a very good thing -- many drugs which may be repositionable have little or no patent life yet, meaning companies will find it difficult to invest in them with any hope for a return.
Gene Logic / Ore has one repositioned drug which has gone through clinical trials, GL1001 (nee MLN4760). This is a drug originally developed by Millennium as an ACE2 inhibitor. Since I'm among the discoverers of ACE2, I tend to key an eye on this one. Millennium gave it a whirl in obesity, but now Gene Logic has found a signal in inflammatory bowel disease in animal models.
That Pfizer bothered to file a patent is significant, as it triggered a milestone payment -- amount unspecified, but these are usually something interesting. But that is still a long way from starting a new trial -- that will be the real milestone, and whichever drug repositioning firm can claim that will really be crowing -- that is, until somebody actually gets a drug approved this way.
Friday, November 16, 2007
Docs eager to winkle wrinkles, slow to hole a mole
I'm no fan of the local TV news broadcasts & therefore rarely catch them. So it was quite by accident that I caught a story last night that is the sort to give one the shudders.
The station had two staffers call 20+ dermatologists. One staffer would state that she had a suspicious mole to be checked out, whereas the other staffer would call requesting an appointment for cosmetic Botox. Now, at some level the results shouldn't be surprising, as if the pattern was the opposite it wouldn't have made the local news. But what was striking was the range of difference: in one case the mole would get an appointment several months in the future, but the same office would be willing to Botox away the next day. Yikes!
Perhaps more striking was the one doc who showed such a pattern who was interviewed on camera. She made no apologies and showed neither shame nor remorse. Her practice is no longer taking on new 'medical' patients, but is happy to accept new cosmetic ones. She did say that perhaps if the caller had been more persistant, perhaps they would have gotten a sooner appointment. Alas, the interviewer did not ask her to explain the ethics of the situation. It is not hard to imagine that many patients calling about a mole are on the fence as to whether to worry or not, and being given a long wait will push them back to complacency (hey, if the doctor's not worried about it why should I be?). Some small fraction of those persons may have early stage melanomas, with potentially lethal results from delay in removal.
It's not hard to guess at the driver of this issue: Botox is elective, probably not covered by insurance, and therefore patients will pay top dollar; mole screening is covered & governed by the immense pricing power of insurance companies.
Somewhere in the last week or so I saw an article commenting that increasing numbers of doctors from a wide variety of specialties are performing cosmetic procedures. A Polyanna might think this would provide the competition to drive the dermatologists back to doing the important stuff, but more likely the corruption will just spread to more specialties. In high school I once switched dentists because it was impossible to get appointments at my long-time dentist, but I went running back after a few appointments when I realized the new guy opened every session with an inquiry as to whether I might want to change out my cap for a better color match.
A real pessimist might note that these new-fangled genetic tests are coming down the pike, that they may also be considered elective and not covered by insurance, and may represent another monetary siren tempting docs to neglect treating disease.
By coincidence, I had just watched the one movie I know of that opens with a monologue on ethics, Miller's Crossing. Caspar & the interviewed doc would be unlikely to have any arguments in that department.
The station had two staffers call 20+ dermatologists. One staffer would state that she had a suspicious mole to be checked out, whereas the other staffer would call requesting an appointment for cosmetic Botox. Now, at some level the results shouldn't be surprising, as if the pattern was the opposite it wouldn't have made the local news. But what was striking was the range of difference: in one case the mole would get an appointment several months in the future, but the same office would be willing to Botox away the next day. Yikes!
Perhaps more striking was the one doc who showed such a pattern who was interviewed on camera. She made no apologies and showed neither shame nor remorse. Her practice is no longer taking on new 'medical' patients, but is happy to accept new cosmetic ones. She did say that perhaps if the caller had been more persistant, perhaps they would have gotten a sooner appointment. Alas, the interviewer did not ask her to explain the ethics of the situation. It is not hard to imagine that many patients calling about a mole are on the fence as to whether to worry or not, and being given a long wait will push them back to complacency (hey, if the doctor's not worried about it why should I be?). Some small fraction of those persons may have early stage melanomas, with potentially lethal results from delay in removal.
It's not hard to guess at the driver of this issue: Botox is elective, probably not covered by insurance, and therefore patients will pay top dollar; mole screening is covered & governed by the immense pricing power of insurance companies.
Somewhere in the last week or so I saw an article commenting that increasing numbers of doctors from a wide variety of specialties are performing cosmetic procedures. A Polyanna might think this would provide the competition to drive the dermatologists back to doing the important stuff, but more likely the corruption will just spread to more specialties. In high school I once switched dentists because it was impossible to get appointments at my long-time dentist, but I went running back after a few appointments when I realized the new guy opened every session with an inquiry as to whether I might want to change out my cap for a better color match.
A real pessimist might note that these new-fangled genetic tests are coming down the pike, that they may also be considered elective and not covered by insurance, and may represent another monetary siren tempting docs to neglect treating disease.
By coincidence, I had just watched the one movie I know of that opens with a monologue on ethics, Miller's Crossing. Caspar & the interviewed doc would be unlikely to have any arguments in that department.
Thursday, November 15, 2007
Trapping the wily VEGF
A significant challenge in pharmacology is the correct dosing of the drug. "The dose makes the poison" is an adage in toxicology, but "the dose makes the drug" just as much. Too little drug and insufficient effect will occur; too much and the patient is likely to suffer from toxic side-effects.
Traditionally, drug dosages evolved completely empirically. Many drugs have profiles allowing very crude dosing -- "take two aspirins and call me in the morning" is remarkable advice, remarkable because it generally works. Other drugs, particularly in trials, are dosed by body size. This makes rough sense, as if you wish to obtain a certain concentration of drug the amount of body it will be diluted in should be taken into account.
Over time various influences on dosing have been realized. I ate a grapefruit today pondering whether one day this small pleasure will be forbidden to me; the metabolism of many drugs is altered by natural compounds present in grapefruit. Individual variation plays a major role as well, with some chemotherapy drugs at normal doses near-lethal to small fractions of the population, because those individuals metabolize the drug differently. Some drugs have notoriously narrow dosing windows: underdose a heart patient and they may have angina or other nasty events; overdose them and they can have nosebleeds which simply won't end.
It is hard enough to dose drugs for which there are decades of experience or which are relatives of drugs with long pedigrees. Dosing brand new agents with new activity profiles is far more difficult. Hence, there is a real need for compasses which could point the way to the correct dose.
VEGF is an important soluble signaling factor which stimulates angiogenesis, the formation of new blood vessels. Anti-angiogenesis agents have emerged as an important tool in oncology and also in the vision-robbing disease macular degeneration. VEGF can be targeted in a number of ways: the antibody drug Avastin (bevacizumab) directly binds VEGF, whereas multi-targeted ("dirty") kinase inhibitors such as Nexavar (sorafanib) and Sutent (sunitinib) knock out the cellular receptors for VEGF (which are tyrosine kinases) among their many targets.
VEGF-Trap is an investigation drug being developed by Regeneron, one of those feline biotech companies (9 lives!) which keep plugging along. VEGF-Trap is a pastiche of carefully chosen protein parts: pieces of two different human VEGF receptors plus a bit from a human antibody (IgG1) constant region.
In a new paper in PNAS (open access) the Regeneron folks show that VEGF-Trap forms stable, inert, monomeric complexes with VEGF which remain in circulation. By measuring the amount of free and VEGF-complexed VEGF-Trap in circulation they can measure VEGF levels and identify a dose which ensures that maximal trapping occurs. If insufficient drug is applied, then little or no free VEGF-Trap is detected.
One significant surprise, in both mice and humans, is that VEGF levels are higher than previously reported. Furthermore, VEGF levels do not differ greatly between individuals with cancer (human patients or xenografted mice) and those without. Human and mouse VEGF levels were very similar, when normalized for body mass. Maximal anti-tumor effects were observed in the mouse models at the dosing point where free VEGF-Trap was observed, suggesting that this method of VEGF measurement can guide dosing.
Can you do the same trick with bevacizumab? Not according to the paper: antibodies form multivalent complexes with their targets, and these complexes are removed from circulation by various mechanisms. Measurements of bound complex are therefore difficult and not informative.
During my previous job I got interested in whether VEGF, or other angiogenic mediators, might be useful for patient stratification. Several papers claimed that soluble angiogenesis factor levels were useful in predicting cancer outcome, but when I compared the measurements in the papers they weren't even on the same scale: the reported baseline measurements in normal individuals were completely different. It didn't invalidate the concept, but certainly prevented any useful synthesis of various papers.
John S. Rudge, Jocelyn Holash, Donna Hylton, Michelle Russell, Shelly Jiang, Raymond Leidich, Nicholas Papadopoulos, Erica A. Pyles, Al Torri, Stanley J. Wiegand, Gavin Thurston, Neil Stahl, and George D. Yancopoulos
VEGF Trap complex formation measures production rates of VEGF, providing a biomarker for predicting efficacious angiogenic blockade
PNAS published November 13, 2007, DOI: 10.1073/pnas.0708865104
Wednesday, November 14, 2007
Ring around the protein...
One of the journals I monitor by RSS is Nucleic Acids Research, and the usual steady flow of new items has become a torrent, mostly about databases. Yes, the annual Database Issue is on its way and the Advance Access shows the signs. And, it's all Open Access.
Every year this issue grows and grows, and each year I skim through all the little niche databases. They may be small and esoteric, but somebody has a passion for that niche & that's great!
I've always liked oddities & anomalies in biology: the rules are useful, but the mess is fascinating. Somewhere in my undergraduate days I came across the fact that there were known examples of circularly permuted proteins, proteins whose final sequence is attained by moving a segment from the tail end around to the front. But somehow the existence of proteins whose mature form is a circle (via a post-translational step) had escaped me. But now that void is filled, as I can loop around to CyBase, a database of cyclic proteins and peptides.
Why circlets? Well, one obvious advantage is two fewer free ends for proteases to make mischief of -- and many of these proteins are protease inhibitors. Indeed, the stability extends to other abuses, with the suggestion that these might make interesting scaffolds for drug development. Circles also make for attractive sequence profile displays. And not only does the database cover naturally cyclic proteins, but has tools to help you design your own!
Conan K. L. Wang, Quentin Kaas, Laurent Chiche, and David J. Craik
CyBase: a database of cyclic protein sequences and structures, with applications in protein discovery and engineering
Nucleic Acids Research Advance Access published on November 5, 2007.
doi:10.1093/nar/gkm953
Monday, November 12, 2007
Just how wrong is Marilyn vos Savant?
Marilyn vos Savant is a writer of a regular column in Parade magazine. These columns address many things, but often have interesting logic puzzles. Given that she is claimed to have the highest recorded IQ ever, whole sites have sprung up to find fault with her writings. Now, I'll confess I'm always looking for an angle -- and rarely finding one.
But this past Sunday, she gave me a bit of an opening: in response to a question as to whether there are any beneficial viruses. Her response:
The first two sentences and most of the third are dead on the money: bacteria are unicellular organisms, viruses aren't considered alive & invade other cells, where they can lie dormant or immediately go crazy. However, that last bit is the clincher. Apparently Ms. vos Savant is unaware that in the bacterial world there are examples of viruses benefiting their host by bringing along useful genetic stuff. Diptheria is one example, in which the toxin (which presumably helps the bacterial host) is encoded by a virus (phage).
Are there examples outside of bacteria? I don't know of any, but I'm hardly up on my viruses. Moreover, how would we know? Suppose there were viruses which were simply neutral (or nearly so), would we have ever detected them?
Also, in a broader sense some of those phage out there may be an important ecological control on bacterial nasties. So this could be another class of "beneficial" viruses.
Just because you are a parasite doesn't mean you're guaranteed bad!
But this past Sunday, she gave me a bit of an opening: in response to a question as to whether there are any beneficial viruses. Her response:
No. Bacteria are living one-celled microorganisms. By contrast, viruses aren’t alive in a strict sense: They are the ultimate parasites and cannot replicate without a host. They invade the cells of animals, plants and even bacteria, then either lie dormant or get into the driver’s seat and cause changes that disrupt normal cell functioning, the very essence of disease.
The first two sentences and most of the third are dead on the money: bacteria are unicellular organisms, viruses aren't considered alive & invade other cells, where they can lie dormant or immediately go crazy. However, that last bit is the clincher. Apparently Ms. vos Savant is unaware that in the bacterial world there are examples of viruses benefiting their host by bringing along useful genetic stuff. Diptheria is one example, in which the toxin (which presumably helps the bacterial host) is encoded by a virus (phage).
Are there examples outside of bacteria? I don't know of any, but I'm hardly up on my viruses. Moreover, how would we know? Suppose there were viruses which were simply neutral (or nearly so), would we have ever detected them?
Also, in a broader sense some of those phage out there may be an important ecological control on bacterial nasties. So this could be another class of "beneficial" viruses.
Just because you are a parasite doesn't mean you're guaranteed bad!
Wednesday, November 07, 2007
Can you Flex your DNA?
One component of many employment benefit packages in the U.S. (I don't know about other countries) is a Medical Flexible Spending Account. At the beginning of the plan year you choose an amount to set aside from each paycheck which goes pre-tax into an account, the contents of which can be used to reimburse medical expenses. Depending on your tax bracket, this is equivalent to getting a 20-35% discount on your medical bills.
These accounts show the many fingerprints of bureaucracy. Contribution limits are stiff, often both a percentage of pay and an absolute maximum (current $4K, I think). There are, as one might expect, curiosities to the restrictions on what the money can be used for. Health insurance premiums are not coverable, but co-pays and co-insurance are. Both the basic eyeglass frame and the Elton John special are generally reimbursable. A periodontal visit is reimbursable, but the dental floss which might prevent or moderate that visit is not. In general, you must save all your receipts and then send them in to the plan operator. Some offer a handy debit card that works with one of the credit card networks, but if you spend on anything out-of-bounds you'll need those receipts to sort it out -- and you may need those receipts in any case. Lots of paper to save.
But perhaps the most burdensome requirement is use-it-or-lose-it; any money not spent by the end of the plan year is forfeited to the plan operator. Over the last year or so the IRS has loosened this restriction to allow some overlap in plan years, but not all plans allow such. So, you must carefully plan your expenses or the whole benefit is lost -- or worse. And, should you wish a mid-course correction, that's generally not allowed -- you can't change your contribution event unless something big happens, such as the addition or subtraction of a family member (the former is hard to plan precisely, the latter should never be planned!).
So, around this time companies offering covered items start urging folks to check their account balances and spend them before they lose them. Eyeglass merchants are at the head of the line, but so are laser vision correction places.
Which leads to the title question: can you use MedFlex account funds to pay for DNA testing? I honestly don't know, and really nobody lacking a tax law specialty has any business answering the question. But, if I were running one of the personal genomics startups out there I'd be finding out the answer. Perhaps a precedent has been set for other diagnostic procedures not (yet?) well recognized to be of medical value, such as the full-body scans which were heavily marketed a bunch of years back. For if these funds are available, then that is a ready pot of money which might be spent. One big ticket receipt sure wouldn't be a pain to submit, and if the companies were clever they could split the bill over two fiscal years (say one covering the scan and one the counseling) to enable two plan years to be charged with the expense. I don't know whether it would make good medical sense to have such a scan, but some folks on fence might be swayed if they could see it as a bargain.
These accounts show the many fingerprints of bureaucracy. Contribution limits are stiff, often both a percentage of pay and an absolute maximum (current $4K, I think). There are, as one might expect, curiosities to the restrictions on what the money can be used for. Health insurance premiums are not coverable, but co-pays and co-insurance are. Both the basic eyeglass frame and the Elton John special are generally reimbursable. A periodontal visit is reimbursable, but the dental floss which might prevent or moderate that visit is not. In general, you must save all your receipts and then send them in to the plan operator. Some offer a handy debit card that works with one of the credit card networks, but if you spend on anything out-of-bounds you'll need those receipts to sort it out -- and you may need those receipts in any case. Lots of paper to save.
But perhaps the most burdensome requirement is use-it-or-lose-it; any money not spent by the end of the plan year is forfeited to the plan operator. Over the last year or so the IRS has loosened this restriction to allow some overlap in plan years, but not all plans allow such. So, you must carefully plan your expenses or the whole benefit is lost -- or worse. And, should you wish a mid-course correction, that's generally not allowed -- you can't change your contribution event unless something big happens, such as the addition or subtraction of a family member (the former is hard to plan precisely, the latter should never be planned!).
So, around this time companies offering covered items start urging folks to check their account balances and spend them before they lose them. Eyeglass merchants are at the head of the line, but so are laser vision correction places.
Which leads to the title question: can you use MedFlex account funds to pay for DNA testing? I honestly don't know, and really nobody lacking a tax law specialty has any business answering the question. But, if I were running one of the personal genomics startups out there I'd be finding out the answer. Perhaps a precedent has been set for other diagnostic procedures not (yet?) well recognized to be of medical value, such as the full-body scans which were heavily marketed a bunch of years back. For if these funds are available, then that is a ready pot of money which might be spent. One big ticket receipt sure wouldn't be a pain to submit, and if the companies were clever they could split the bill over two fiscal years (say one covering the scan and one the counseling) to enable two plan years to be charged with the expense. I don't know whether it would make good medical sense to have such a scan, but some folks on fence might be swayed if they could see it as a bargain.
Tuesday, November 06, 2007
Lung Cancer Genomics
A large lung cancer genomics study has been making a big splash. Using SNP microarrays to look for changes in the copy number of genes across the genome, the group looked at a large batch of lung adenocarcinoma samples. Note: the paper will require a Nature subscription, but the supplementary materials are available to all.
As with most such studies, there was some serious sample attrition. They started with 528 tumor samples, of which 371 gave high-quality data. 242 of these had matched normal tissue samples. All of the samples were snap-frozen, meaning the surgeon cut it out and the sample was immediately frozen in liquid nitrogen.
The sub-morphology of the samples is surprisingly murky; much of the text focuses on Non-Small Cell Lung Cancer (NSCLC), the most common form of lung adenocarcinoma, but the descriptions of the samples do not rule out other forms.
After hybridizing these to arrays, a new algorithm called GISTIC, whose full description is apparently in press, was used to identify genomic regions which were either deficient or amplified in multiple samples.
Many changes were found, which is no surprise given that cancer tends to hash the genome. Some of these changes are huge: 26 recurrent events involving alteration of at least half a chromosome arm. Others are more focused.
One confounding factor is that no tumor sample is homogeneous, and in particular there is some contamination with normal cells. These cells contribute DNA to the analysis and in particular make it more difficult to detect Loss-of-Heterozygosity (LOH), in which a region is at normal copy number but both copies are the same, such as both carrying the same mutated tumor suppressor.
Seven recurrent focal deletions were identified, two of which cover the known tumor suppressors CDKN2A and CDKN2B, inhibitors of the cell cycle regulatory cyclin-dependent kinases. The corresponding kinases were found in recurrently amplified regions; an neat but evil symmetry. Tumor suppressors PTEN and RB1 were found also in recurrent deletions. The remaining recurrent deletions hit genes not well characterized as tumor suppressors. One hits the phosphatase PTPRD -- the first time such deletions have been found in primary clinical specimens. Another hits PDE4D, a gene known to be active in airway cells. A third takes out a gene of unknown function, AUTS2.
In order to gain further evidence that these deletions are not simply epiphenomena of genomic instability, targeted sequencing was used to look for point mutants. Only PTRPD yielded point mutants from tumor samples, several of which are predicted to disable the enzymatic function of this gene's product.
On the amplification side, 24 recurrent amplifications were observed. Three cover known bad actors: EGFR (target of Iressa, Tarceva, Erbitux, etc), KRAS and ERBB2 (aka HER2, the target for Herceptin). Another amplification covers TERT, a component of the telomerase enzyme which is required for cellular immortality, a hallmark of cancer. Another amplification covers VEGFA, a driver of angiogenesis and part of the system targeted by drugs such as Avastin. Other amplifications, as mentioned above, target cell cycle regulation: CDK4, CDK6 and CCND1.
The most common amplification has gotten a lot of press, as it covered a gene not previously implicated in lung cancer: NKX2-1. A neighboring gene (MBIP2) was present in all but one of the amplifications, and so NKX2-1 was focused on. Fluorescent In Situ Hybridization (FISH), a technique which can resolve amplification on a cell-by-cell basis in a tissue sample, confirmed the frequent amplification of NKX2-1 specifically in tumor cells. Resequencing of NKX2-1, however, failed to reveal any point mutations in the tumor samples. RNAi in lung cancer cell lines with NKX2-1 amplification showed a reduction of a commonly-used tumor-likeness measure (anchorage-independent growth). This effect was not seen in a cell line with undetectable NKX2-1 expression, nor was it detected when MBIP2 was knocked down. Previous knockout mouse data has pointed to a key role for NKX2-1 in lung cell development. The protein product is a transcription factor, and the amplification of lineage-specific transcription factors has been observed in other tumors.
What will the clinical impact of this research be? None of the targetable genes which were amplified are novel, so this will nudge interest further along (such as in using Herceptin in select lung cancers), but not radically change things. Transcription factors in general have no history of being targeted with drugs, so it is unlikely that anything will come rapidly from the NKX2-1 observations. On the other hand, there will probably be a lot of work to try to characterize how NKX2-1 drives tumor development, such as to identify downstream pathways.
At least some of the press coverage has remarked on the price tag for this work & the surrounding controversy over the Cancer Genome Project that this represents. The claimed figure is $1 million, which does not seem at all outrageous given the large number of microarrays used (over one thousand, if I'm adding the right numbers) -- a few hundred dollars per microarray for the chip and processing is not unreasonable, and the study did a bunch more (analysis, sequencing, RNAi). If such a study were to be repeated at todays prices in the next 5 big cancer killers (breast, ovarian, prostate, pancreatic, colon), it means another $5M not spent on other approaches. In particular, the debate centers around whether the focus should be on more functional approaches rather than genomics surveys. As fond as I am of genomics approaches, it is worth pondering how else society might spend these resources.
It is also worth noting what the study didn't or couldn't find. A large number of known lung cancer relevant genes did not turn up or turned up only weakly. In particular, p53 is mutated in huge numbers of cancers but didn't really turn up here. The technique used will be blind to point mutants and also can't detect balanced translocations. Nor could it detect epigenetic silencing. If you want to chase after those, then it is more genomics -- which is probably one of the things that eats at critics, the appearance that genomics will never stop finding ways to burn money.
Weir et al. Nature Advance Publication. Characterizing the cancer genome in lung adenocarcinoma. doi:10.1038/nature06358
Wednesday, October 31, 2007
Scientific Easter Eggs
Tonight, of course, is Halloween, one of the many holidays which in the U.S. has a serious sweet tooth. After taking Version 2.0 around for the tradition of gentle extortion on this day, I indulge in my own rituals -- listening to Saint-Saens & reading The Raven. It isn't exactly the right time, but the confectionery angle got me thinking about other sweet holidays, and then to Easter Eggs -- of the scientific kind.
There was a recent complaint in Nature about the growing shift of information from the printed versions of articles to the Supplementary Online Material (SOM). I can definitely sympathize -- as the writer complained, key details have been migrating to the SOM, meaning that sometimes you can't read the print version and really tackle it scientifically. In particular, Materials & Methods sections of many papers have been eviscerated, with the key entrails showing up in the SOM. Most of the point of my print subscriptions to Science & Nature is to be able to read them during my Internet-free commute. Worse, the SOM becomes an appendage in danger of being lost or misdirected -- such as in a recent manuscript I reviewed which showed up without the supplements.
For better or worse, editors & authors have shared interest in shifting things from print to the SOM. For editors, online is cheap. For authors, it is a way to cram more in to fixed paper size limits. Clearly some material (such as videos) can only go into SOMs, and lots of supporting data really does belong there.
In computer code, an Easter Egg is a hidden surprise -- if you know the right combination of keystrokes or commands or such, something interesting (and generally irrelevant to the program) will show up. I'm not sure I've actually ever seen one -- I'm generally too impatient to deal with such things, but I do recognize they exist. Granted, perhaps some of that programming effort would be better spent wringing a few more bugs out, but it is a way for coders to blow off steam.
I propose that a scientific Easter Egg is the inclusion in Supplementary Online Material of valuable scientific data which is peripheral to the main thrust of the paper, but is nevertheless a significant advance. Such events are probably rare, as it requires a certain mindset to bury a possible Minimal-Publishable-Unit in another paper's SOM, but on the other hand it beats something never being published -- and perhaps it is interesting to some but viewed as too minor to merit a paper.
I'll give you an example, from the Church lab. George has long been burying stuff in papers -- for example, one of the footnotes to the original multiplex sequencing paper declared that the technology was being used to shotgun sequence Salmonella typhi AND Escherichia coli! Alas, the project was ahead of the technology & never completed. But a much better Easter Egg is in the first large-scale polony sequencing paper (PDF ; SOM). Supplementary Figure 2 is really an in-depth study of the site preferences of the Type IIS restriction enzyme MmeI -- driven by about 20K of sequencing examples. This is really a bit of restriction enzymology hiding in a sequencing paper. Because the enzyme is used in the method, it is relevant -- but not quite critical. The enzyme preferences are important because it could create biases in sequence sampling, but it is hardly the main point of the paper -- which is why it is in the SOM.
I'm sure there are even better examples out there. What is the most interesting tangential information you have seen in an SOM?
Shendure, J, Porreca, GJ, Reppas, NB, Lin, X, McCutcheon, JP, Rosenbaum, AM, Wang, MD , Zhang, K, Mitra, RD, Church, GM (2005) Accurate Multiplex Polony Sequencing of an Evolved Bacterial Genome Science 309(5741):1728-32. DOI: 10.1126/science.1117389
Church, G.M., and Kieffer-Higgins, S. (1988) Multiplex DNA sequencing . Science 240: 185-188. DOI: 10.1126/science.3353714
Subscribe to:
Posts (Atom)