Monday, January 31, 2005
Friday, January 28, 2005
Presidential inauguration funnies
Thursday, January 27, 2005
Semi-random thesis thoughts
While these cell-to-cell communication circuits are cool and all, they suffer a bit from the fact that they're basically serial communication ie they're a bit like having a conversation where you can only answer "yes" or "no" to questions -- you eventually get to the point, but it takes a while. I've been thinking about making this a more powerful system such that you can actually exchange whole "sentences", so to speak. I have a decent idea, on paper, how to do this but I suspect that actually trying it in the lab would become painful pretty quickly, especially given my lack of lab experience. It's a project that I'll probably keep hammering away at during my current rotation with Tom Knight, but I'm also thinking about other stuff.
The other area I'm getting interested in developmental biology, which is concerned with how you make a whole critter when all you've got to at the beginning is a single cell. On the face of it, it's just insane that you can pack the entire blueprint for something as big as an elephant, as weird as a duck-billed platypus [a mammal that lays eggs yet suckles its young after hatching] or as, ahem, "smart" as GW Bush into a single cell that you can't even see without a microscope.
There are oodles of observational data [ie what happens] about how certain model organisms [like the fruitfly, a particular species of frog, the mouse, humans etc] develop, as well as lots of data about what individual genes are involved in developing the adult body plan. However, so far there are fairly few large-scale "wiring diagrams" that tie all the bits involved together and really give a good global explanation of how & why you get from the fertilized egg to the various lifecycle stages. The only thing I've been able to find that comes close is some work out of Caltech, by Eric Davidson's group, which has mapped out this sort of diagram for the sea urchin. [PubMed reference to the paper]. A look at the number of authors on that paper should give you an idea of how much work it is to do this for something as "lowly" as a sea urchin; "higher" animals will probably have an even more complex set of interactions, requiring correspondingly more work, though that remains to be proven.
What appeals to me about this area of biology is that
a) I've always been interested in how you get "emergent" behavior [like a whole animal] out of the interaction of lots of little "dumb" parts [like cells] that are just following fairly simple rules, ad animal development is a really cool example of that
b) there is lots of room for computational work: to attack the problem you need to do large-scale experiments to figure out what genes are active and perform computational analysis of the results, as well as build a computational model of the gene network to explain the progression through the various stages of development. In other words, lots of fodder for a computer geek like me.
So, I'm in the process of reading "Developmental Biology" and whatever papers I can find on PubMed to see whether this is something I want to pursue further. I'm also going to talk to a professor at Harvard, Radhika Nagpal, since this is an area she's interested in [as a bonus, she has good connections to Drew Endy and Tom Knight, having worked with both of them when she was a postdoc at MIT].
Some thoughts about public transport
Observation #1: Subways are less stochastic than buses. Subways are not subject to the vagaries of traffic, don't have to deal with the time penalty of new people getting on and having to pay at each stop and have a pre-determined number of stops. This makes taking a subway much more predictable, and hence, preferable to a bus.
Observation #2: Being on a subway/bus with lots of undergraduates will make you lose brain cells. Our bus route took us past Northeastern University, Wentworth Institute of Technology and probably a couple other of the tertiary educational institutions that seem to sprout out of Boston's soil like mushrooms. As a result, we got to spend lots of time listening to the conversations of 18-21 year-olds and, boy oh boy, are they ever fascinating. Not. I think all conversations we overheard can be reconstructed through pseudo-random recombinations [only loosely constrained by grammar] of the words/phrases "like", "oh my god", "totally", "beer", "bar", "suck", "drink/drunk", "hot", "cute", "class" and "exam". [I'm sure I didn't sound any better at that age, but that doesn't mean I can't find it annoying now.]
Observation #3: Lots of people seem to have no sense of privacy. It's somewhat unfair to single out a particular age group as having inane conversations -- my subway conversations aren't exactly scintillating gems of erudition either. But, and this is a key difference, I try to pitch my voice such that the only person who can hear me is whoever I'm talking to. Lots of people don't seem to have that filter, whether they're talking to somebody face-to-face or on their cellphone -- whether you want to or not, you overhear their conversation. I always wonder whether they don't realize this or just don't care, perhaps being of the opinion that their life is so fascinating that of course everybody would want to hear about it.
Observation #4: Music is insulation. I'd never really thought about why so many people seem to be listening to music on the subway or bus, but today it struck me that it's a form of insulation. You don't have to deal with hearing other people's conversations, can control what you hear and it provides a way to stay entertained while your mode of transport trundles ever so slowly towards its destination. I may have to start listening to music too.
All this brings me to a few proposals:
Proposal #1: Each subway car should be outfitted with a decibel meter that indicates the noise level inside the car. This would allow passengers to avoid getting into a car that contains noise-generating entities like rambunctious teenagers and young men who want to show everybody how tough they are and hence spend lots of time cursing loudly.
Proposal #2: Somewhat in line with proposal #1, there should be kids [up to 18], young adults [18-25] and adult [25+] sections on public transportation. This would prevent friction due to incompatible population segments.
Proposal #3: Public places, like the aquarium or zoo, should give each set of visitors helmets with a built-in microphone and radio system, tuned to a frequency unique to that set of visitors. This would allow each group to talk to each other but not have to hear everybody else.
Ok, so I'm a curmudgeon who really likes undisturbed private space. Maybe I should just build myself a Fortress of Solitude [on a tropical island, though, not in the Arctic] and retreat to it.
Wednesday, January 26, 2005
Walking uphill both ways, through the snow
After the 2+ feet of snow we got a couple of days ago, we're now getting another 6+ inches. I'm finally at the point where my reaction is no longer "Ooh, snow is cool, it makes everything look pretty" but rather "#%#$@#$, more #$#$ snow ?". And that's without even having to shovel any snow. I think if I had to do that, I'd just buy a snowmobile and forget about shoveling anything.
... and, caused by the snow, another entry in the category of "Situations you never thought you'd find yourself in": as I mentioned earlier, the pipes in our apartment building froze after the blizzard, so we didn't have any water in our kitchen. After calling our landlord in vain for a day [smart man, he probably turned off his cellphone], we finally reached him on Monday. A bit of explanation is in order about him: he's a Portuguese man who apparently came to the US about 20 years ago with just the clothes on his back and now owns several apartment buildings. In other words, he works pretty damn hard. However, and this is the downside of it, he still insists on doing pretty much all the necessary maintenance on his apartments himself. Or, more accurately, I should say that he and his family [his wife, five daughters, one son] do all the maintenance -- I've seen his wife shoveling snow outside our apartment a few times now and one of his kids sweeps the hallways in our apartment building every Saturday morning at 7 am or some such ungodly hour. What this friends-and-family maintenance plan means is that the maintenance is generally pretty sketchy because his mode of operation is pretty much "patch it up for now so the tenants stop complaining, don't think too hard about the long term".
In the case of our frozen pipes, he arrived on Monday afternoon with two of his teenage daughters in tow and disappeared into the basement. Christina and I went down there to check on him around 4pm and he was in the process of thawing the frozen pipes [via some gadget that, I think, generates heat in the pipes by running a current through them ...] and boldly proclaimed that he'd be done in about an hour or so. When four hours had gone by and we still didn't have water, I went down there to check on what was going on [at Christina's, uhm, request ;-)] and found him soldering water pipes together with just a little itty-bitty flashlight providing the illumination [there's no light in the basement] and a flame and spool of soldering wire that seemed more appropriate for use on a high school shop project than fixing plumbing for a 6-unit apartment building. I went back upstairs to get another flashlight and then found myself helping a 50+ year-old Portuguese man, his three teenage daughters and teenage son to solder together pipes in a damp, dark basement. Talk about the blind leading the blind ... If you'd shown me that scene a year ago and asked me to spin a story about how I could possibly have gotten there, I don't think I would have been able to come up with anything remotely plausible. Life is funny that way.
Sunday, January 23, 2005
This is pure snow ! Do you know what the street value of this neighborhood is ?
Christina and I are experiencing our very first nor'easter and, boy, it's dumping quite a bit of snow. We're at over 2 feet and rising. And the howling wind [gusts of over 50 mph, apparently] makes me all the more grateful that I'm inside and don't need to leave the house for the next few days; I hope all the homeless people managed to make it into a shelter.
It looks like the pipes supplying water to our kitchen tap are frozen -- when you turn on the tap, you get a big fat lot of nothing. Doing some spelunking in the basement didn't reveal much other than a pipe next to a broken window that appeared to be steaming but we couldn't really investigate any further because it's in a locked section of the basement. At least the water is still running in the bathroom and we have power; let's hope that continues to be the case.
Pity the snow can't be put to better use, like snowboarding. It'd be perfect for it -- it's light, dry and fluffy, just the right sort of thing for throwing up roostertails. Meanwhile, my old stomping ground, the Mt. Baker Ski Area, is hurting for snow.
Let a thousand pictures bloom
Saturday, January 22, 2005
Lifestyles of the bizarre and insectile
Friday, January 21, 2005
My new foreign policy: lots of dove, very little hawk
Now, all this talent makes for a great workout atmosphere because there's lots of energy and enthusiasm during the workout. On the flip side, that much concentrated ass-kicking-ability can be problematic when it's your ass. I went back to class yesterday after a 6-week layoff and walked into a workout that consisted mainly of sparring because a lot of people are preparing for a tournament that's next weekend. Now, let's play a little Jeopardy. Here is your clue, from the category "Feel the beat":
"This is what you get when you put an over-the-hill, out-of-shape fighter [ie Alex] up against somebody who is younger, faster, in excellent shape and just plain better"
If you said "What is a beating ?", you are correct, 'coz that's what I took yesterday. We did a bunch of round-robin sparring [ie sparring with lots of different partners] and all I can say is that kids these days have no respect for their elders. That, and "Ouch, that hurts. Ouch, that hurts too". I'm pretty much a solid mass of bruises from the waist down. I'd be bruised up top too, but luckily I was wearing a chest protector, or else I'd have footmarks all over my chest and ribcage. Now that's partially my fault -- I kept asking these guys to spar me, thinking "That last shot they got was a lucky shot, that won't happen again ... arrrgh, *cough*, that was lucky too ... c'mon bring it ... ouchh....". You get the idea. In retrospect, it must have been a bit like the Black Knight in Monty Python's "Holy Grail" who insisted that having his arms chopped off was "just a flesh wound".
I learned a very important lesson yesterday: at my advanced age and stage of decrepitude, I should follow a foreign policy based on cooperativity and non-aggression towards people that have approximately the same dimensions as me. In other words, start picking on much smaller people ;-)
Wednesday, January 19, 2005
Staying in touch with what's "going on" [so to speak]
It's funny and uncomfortably accurate at the same time, in the sense that parents who want to be in any way conversant with what their kids are talking about probably have to spend a fair bit of time actively seeking out the media outlets their kids use. Now you could argue that that's always been the case ie that kids have always gotten their "news" and topics of conversation from things that adults don't know much about, but I think the disconnect is probably getting worse because there are so many other outlets [or inlets, as the case may be ...] now, and they're multiplying quickly. That said, maybe the current generation of parents is technically savvy enough to be able to also adjust quickly to the rapid emergence of new forms of media. As usual, only time will tell.
Sunday, January 16, 2005
Yeah, less thought is just what we need
These two products come from different eras - the book from the prehistoric world before silicon, and the music player from five minutes ago - but both suggest to consumers that there is a way to remain thinking, feeling people in a world overgrown with data, options and demands, said David Bennahum, who writes about technology issues for the online magazine Slate and for Wired magazine.
"They are two things that say your rational process of making sense of things is a model that may be obsolete," he said. " 'Life is random' is a really great way of shrugging your shoulders in a Buddhist way of nonattachment."
Ah, yes, the process of reasoning rationally is obsolete because, well, it's just too hard, and leads us to conclusions we don't like, so let's just not do it. Because if we just ignore it, it'll go away, right ? That's the dumbest thing I've heard in a long time. We need more people who think more, not less, and who demand the same from others. Otherwise, we get nonsense like "evolution is just a theory, no better and no worse than creationism" [Panda's Thumb is a good outpost to watch the battle against that particular brand of insanity] and an administration that just suppresses science it doesn't like.
Education isn't the same as rationality, but like the man said: if you think education is expensive, try ignorance.
I suspect, for no reason I can clearly enunciate, that minorities tend to gravitate towards fields that are "hot" [like computer science was/is, and biotech is now] later than other segments of the population ie the uptick in their enrollment tends to come later than that for say, white, Indian or Chinese entrants, and that part of this is due to a certain amount of risk averseness. In other words, there's kind of a "let's wait and see whether this field pans out" attitude. If that theory is true, we should see a spike in black/Hispanic/etc participation in life sciences in a few years, assuming biotech sustains itself and becomes a relatively low-risk, decent-reward career choice [in terms of job security, compensation etc] like engineering currently is. Guess we'll have to wait and see.
I also had an interesting conversation with the CTO of a biological modeling company. While flipping through the attendee list and googling some of them, I found out that he didn't have a PhD, so I was curious whether my understanding that needing a PhD to really have much of a career in biotech was wrong [ie my reaction was along the lines of "Wait, did I get duped into this whole school thing ?" ;-)]. I tracked him down sometime during the day and peppered him with questions for a while, which he graciously answered. Given the fact that he doesn't have a PhD, he was more inclined to discount the importance of one, but even he had to admit that, yes, a PhD still matters in biotech, and will continue to matter for the forseeable future. It turns out that he was one of the founders of the company, which probably helps to explain his current position despite not having the expected "union card". So, at least from that perspective, I'm not totally off my rocker for being in graduate school =)
Friday, January 14, 2005
Systems biology is, like, hard and stuff
All that said, I did get some useful insight into what it's like to do "systems biology" by taking a large data set and trying to analyze it computationally. Basically, I was working on using Bayesian networks [check out the author of the quote at that link -- who knew Michael Jordan was into statistics and graph theory ;-) ?] to analyze some data about how T cells react to certain chemicals. Bayesian networks allow you to come up with statements like "according to this data, A affects B, C and D, but not E; D affects E and G, but not F" and know how likely it is that your statement is correct. [I'm going to refer to such statements as "networks" in the rest of the post, because that's they're networks of interactions.] These sorts of models are nice in that they are human-interpretable ie there are few-enough interactions at each level that a human can look at them, make some sense of them and figure out what experiments to do next.
A lot of my time was spent massaging the data into the right format and getting some existing code to run, but I also did get a chance to generate a bunch of networks. The tricky bit, I found, isn't generating the model, it's figuring out how much you trust it and what to do with it. The problem is that trying to find the absolute "best" network is NP-hard ie it takes an insane amount of computer time. For example, if you only have 4 nodes in the network, there are over 500 possible networks and if you have 10 actors, there are on the order of a billion billion [yes, that's "billion" twice] possible networks. So, since we had about 12 nodes, so we couldn't do an exhaustive search. The only thing to do is follow what's called a "greedy" strategy, which meant making up random networks and trying to optimize them. The more such networks you can generate, the better a chance you have of finding a good one -- it's basically a question of how much computer time you can afford to spend on the problem.
In this case, we didn't have much in the way of computing resources, so we'd run the search for a couple of hours, or overnight, and see what we got, and that's where it became tricky to figure out what to do. Our runs produced networks that didn't look much like the interactions that are known from the actual biological literature ie we had things supposedly affecting each other that were, uhm, "unsupported by the evidence", so to speak. Now, part of the appeal of the sort of "machine learning" procedures like Bayesian networks is that they're supposed to be able to uncover relationships that are there but have never been noticed before, so the question became "Geez, the network we generated says these two things are related, but there's nothing in the literature about it, should we go do some experiments to figure out whether they actually are related ?". In order to answer that question, you have to know how much you trust the generated network ie how likely it is that it's actually showing you something real that's worth investigating further. But, from an abstract perspective, you don't really know how good the network is [ie how much you should trust it] because it could well be that your random search only turned up crappy networks and there are much better ones out there. You could, of course, say that you won't trust any network that doesn't show you what's already known from the literature, but if you'll only trust a network that tells you what you already know, what's the point of doing all this in the first place ?
In our case, the question "Should we trust this network ?" wasn't really that hard to answer -- we knew that we hadn't run the search for nearly long enough and that a lot of the networks we generated were just pure bunk from a biological perspective, so there was no need to rush to the lab bench and try some more experiments. If I had to do this for real, I think the strategy I would adopt would be a combination of using the known biology to restrict the search space somewhat [without totally removing the ability to uncover unknown connections] and running the search for a long time [on the order of weeks, probably].
Of course, all this assumes that the actual data you're using is good ie that everything was measured correctly, that you've actually represented the input data to your algorithm properly etc. If that's not the case, it may well be that you run your search for years and never get anything that makes any sense. Hopefully there are some initial sanity checks you can do before wasting time and money on an extensive computer run.
So, all in all, I definitely learned something about the pains and perils of large-scale biology. My next rotation is going to be back to the small scale [in the figurative sense]: it'll be with Tom Knight [a well-known MIT computer scientist turned biologist], working on cell-to-cell communication in bacteria.
Thursday, January 13, 2005
Amusing tidbits from The Onion
Wednesday, January 05, 2005
Physicists say the funniest things
"The answer is that bacterial viruses make themselves known by the bacteria they destroy, as a small boy announces his presence when a piece of cake disappears."