Sunday, April 30, 2006

Two computer-y items

Item one: The keynote speaker at this weekend's CSBi retreat was Alan Crane, the CEO of Momenta Pharmaceuticals, who bears an uncanny resemblance to Bill Gates. The bit that gave me a "glitch in the Matrix" moment was when he started to talk about monopolies, anti-trust and the FTC. Except that, in this case, he was explaining why having the FTC around was a good thing for his company.

Item two: I received a letter from State Farm Insurance, dated 26th April 2006, to inform me that they were terminating automobile coverage for me as of August 3rd 2004. The only way I can explain this occurrence is that whatever passes for a computer system with them got a little ... behind in what it was doing and is finally getting around to cleaning out some old stuff. Talk about resource starvation ... I have this vision of a thread perpetually relegated to the background, getting thinner and thinner until finally, with its dying breath, it managed to get an old rattling printer, in a musty basement somewhere, to spit out the letter I received.

Saturday, April 29, 2006

42'' x 32'' of postery blandness

That poster I was complaining about having to put together ? Yeah. That. Against much internal resistance [and only because I had to], I finally dragged it out of myself, and am going to hem and haw my way through it today at the annual CSBi retreat. AMY, however, is not featured on it, besides a comment in the Conclusions section about having "constructed strains [that we are] characterizing" because AMY101.1 and AMY201.1 are still terra incognita at this point, only having come into being [accompanied by the obligatory maniacal laughter and triumphant "It's alive !" echoing down the halls] in the last couple of days.

The poster is, as I feared/expected, nothing except a list of "things I'd like to do and why they might be interesting." The bigger problem [ie besides the utter lack of results of any sort] is that I find myself rather ... non-excited by the proposed research described on my poster -- it's basically a bunch of standard DNA hacking to better understand the yeast pheromone response pathway. While that's a perfectly respectable line of research, there's nothing particularly "synthetic" about the biology in it, nor much of a computational aspect, especially of the "machine learning" type that I'd like to do.

I need to fix that, and fast, especially given that my qualifying exam and accompanying thesis proposal are in 2.5 months.

Friday, April 28, 2006

CSBi-cal cancer

Tim Hughes gave a talk yesterday in which he made a comment about a "connection between CSBi and cancer". While he was probably referring to the fact that a couple of the most prominent faculty members that are part of CSBi work [in one fashion or another] on cancer, I couldn't help but think that his statement could also be interpreted as CSBi being a form of cancer slowly taking over MIT, as evidenced by the number of MIT faculty members "associated" with CSBi [where "associated" means "signed up on the CSBi website"].

I suppose it's a reflection of the fact that "systems biology" is the new hotness, despite the fact that nobody can define it [but presumably they know it when they see it...], and everybody wants to be associated with its incandescent light. Or something. In any case, it does make me wonder whether CSBi is becoming so diffuse that there's no point in thinking of it as a distinctive entity anymore. Not that that would necessarily be a bad thing -- as Tozier has pointed out, barriers between disciplines can be annoying. Presumably that'd be all the more the case if a new "discipline" whose guiding motto is "Throw all possible methods at the problem and see which ones stick" [the polite term for this is "interdisciplinary approach", I think] suddenly decided to get all exclusive and clubby in an "You're not interdisciplinary enough" sort of way.

Sunday, April 23, 2006

A vexing question of nomenclature

As I've said before, labeling is important when doing lab work. This is true not just for chemicals, but also for cells -- it's much more convenient to have a short label referring to a particular kind of cell than to always have to say "The cells containing genes X and Y, but without genes A and B ...". The convention [or maybe just -a- convention] in the field of yeast biology when naming cell lines [aka "strains"] created by a particular researcher is to give them a label of the form xxY.### where xx are the researcher's initials, 'Y' denotes the fact that it's a yeast strain and ### is some number, like 123. [Some might argue that this somewhat dry naming scheme, much like the very staid practice of naming yeast genes with a three-letter abbreviation supposed to encode something about their function, reflects the fact that yeast biologists just aren't as much fun as those wild and crazy Drosophila people. But I digress.]

This convention is at the heart of my problem: if I follow it, all my yeast strains will be called ... AMY.

the knee-jerk "Real Male Scientists (TM) don't create anything called 'Amy', they pick names like 'Crusher 11X' or 'BloodHammer 6' or 'RQZ2001.21' " reaction, this notion also disturbs me because I find myself coming up with anthropomorphic questions like:

- Am I creating a line of fembots, with model designations AMY001, AMY002 etc ? [Granted, they'll look much more like this than like that, but still ...]
- Will all the other AMYs be forever jealous of AMY001, because she was my first ?
- Will AMY007 be a yeast secret agent, mating somewhat indiscriminately and generally living a life of ill-repute ? On the plus side, that probably means she'll be impossible to kill ...
- Will AMY067 be a total b!tch, impossible to please, always dying or exhibiting weird behavior at the most incovenient times ?
- Will my cells, in general, be more fickle and inconstant than if they were called something more boring, like, say, RGY ?
- How will Christina take the news that I spend all my time at work with AMY ? Even worse, do I really want my son being told "Daddy can't spend any time with you right now because he's with that !@#$^$%@ AMY" ?

Some possible solutions I've come up with are:
- Buck the trend and put the 'Y' in front of my initials, so my strains will be called YAMxxx. Not great, but a little better.
- Use an initial from one of my middle names, which leads to the names KMYxxx or EMYxxx. Unfortunately, both of those still sound a lot like female names;
it's the "MY" part that's the problem and trading Amy for Kimmy or Emmy doesn't seem like a net win.
- Spell out the letters when talking about my strains ie refer to them as A-M-Y. The drawback here is that it doesn't work in print [and hopefully, at some point, something about my strains will be in print :-)]

... or I could just
accept that, in all likelihood, nobody other than me cares and resign myself to creating a numbered series of AMYs.

Tuesday, April 11, 2006

Don't upset the prisoners, they're very delicate

Apparently an officer at the state's largest prison is going to be disciplined for screening "Brokeback Mountain" for the prison inmates. Money quote:

The problem isn't that the movie is about gay cowboys, but rather the "graphic nature of the sexually explicit scenes," a Massachusetts Department of Correction spokeswoman said.

Ok, so let me get this straight: they're worried about showing sexually explicit scenes. To a) people in prison who b) probably have more gay sex than the protagonists in "Brokeback Mountain" ever dreamed of.

What are they worried about, that the inmates have led such a sheltered life that they'll be traumatized by the sex scenes, or that their "re-education" will be set back ?

I continue to wonder how people can say such patently retarded things with a straight face.

You complete me

..., at least my thesis committee. Sort of. As in: I've finally filled out my thesis committee, except for the one wildcard that's going to be assigned to it by the Powers That Be [aka the CSBi graduate committee]. The members are:

- Fred Winston, a "real" biologist and an authority on all things yeast, especially transcription.
- Ernest Fraenkel, with expertise in investigating transcriptional regulatory networks in yeast and whose lab, in collaboration with others, has generated oodles of genome-wide binding data for yeast transcription factors.
- Alexander van Oudenaarden, who, in addition to being a funny professor, has also done a bunch of really nifty "systems biology" research combining experimental and computational approaches.

.. and, of course, the inimitable Drew.

All in all, I'm very happy with it -- I think I have a good mix of professors with experimental and computational expertise, just what the doctor ordered. Add the fact that they're all very nice and hence liable to get along when put in a room together [ie "play well with others"] and I think I'll get some very good feedback and guidance from them.

"I will lash together a machine of bone and blood ..."

Once again, the good folks at Penny Arcade had me in stitches with their latest comic. It's a perfectly-done pastiche of many of the conventions of the fantasy/horror genre. I especially like the concept of a fear engine.

Tuesday, April 04, 2006

Computational work is starting to look really good

[More whining]
After fighting his way past the lair of contaminated reagents, getting stuck for a week in the swamp of non-growing cells and narrowly winning a long battle with yeastie beasties that refused to eat perfectly good DNA placed in front of them [despite repeated imprecations along the lines of "There are starving cells in the fridge next door ! Eat your DNA !"] , our hapless protagonist now finds himself facing yet another foe: the Plasmid That Is Not What You Think It Is.

Ok, that's enough talking about myself in the third person.

I obtained said plasmid [basically, a bit of DNA] from somebody in another lab and was counting on it to do great things for me, in terms of reducing the amount of work I had to do. Unfortunately, it turns out that there's a large-ish mismatch between what the plasmid is supposed to do [and the DNA sequence it's supposed to contain] and what it actually does. I initially became suspicious when some cells with that plasmid that shouldn't have grown [because I screwed up the experimental procedure] grew like crazy. And my suspicion was further heightened when I looked at the supposed DNA sequence and found it sorely at odds with what it was supposed to be. Further digging revealed that the person I got it from wasn't entirely sure how it had really been constructed and was a bit fuzzy on certain fairly basic aspects of how plasmids work. Not reassuring.

So at this point, I think the safest thing to do is not waste any more time trying to figure out what it actually is, but rather just make my own from scratch. There go another few weeks. *Sigh*

What makes it all worse is that I have to present a poster in ~3.5 weeks at the annual CSBi retreat and I think that pretty much the only thing I'll be able to put on that poster is a fancy-sounding abstract and some high-level ideas that trail off into "... but I don't actually have any results."

At this rate, I may have to acknowledge defeat sooner rather than later and go back to being a keyboard jockey.