You are currently browsing the monthly archive for October 2008.
i hate being this busy. there are demands coming at me from all sides and i’m working at all hours to keep up with them. this week i made a poster, wrote several things for the boss, did my own experiments, put together in various forms my data for the collaborators, gave feedback on ideas and rough drafts, and most importantly (ha) came up with ideas and a start on the lab halloween pumpkin. this is a big deal, we’ve got 4 years of pumpkin championships to maintain.
this makes me a real machine- i see what i’m capable of doing and it’s not difficult. it’s just tiring. i’m sick of going to work first thing and getting home after dark. after these major demands pass, i would like to enjoy a weeknight to myself.
i do need a break, but there’s a meeting this weekend in a neighboring state that i really need to go to. i’m looking forward to the meeting, it’s a great group of people in my field- and a road trip would be kinda nice right now. after work i checked my tires and the spare, but it was too dark outside to peek under the hood. i’m not worried at all about something going wrong, but i always check before road trips out of habit.
best yet is mileage in my car. gas costs far, far less than the mileage rate. heh. joke’s on THEM.
if i can get away with it, i’m so not going to work on monday. a girl needs a day off once every couple of weeks.
today was great evidence this law exists.
the video camera quit working this morning after i had started the pre-experiment injections. punching the cinder block wall several times was probably not needed, but it made me feel better at the time. i am kinda regretting that right now though- ouch. i started my day far later than anticipated.
then my second to last rat exploded on me when i went to inject him (no, not literally) and i stuck myself in a very painful place. fortunately, the contents of the syringe are not harmful- just the needle part.
and finally, i brought home some FREE FOOD today (the ultimate mission of graduate students is to acquire as much free food as possible, for the uninitiated) and i cut my hand open while slicing it! damnit!
i spent my drive home hoping desperately that i did not run out of gas on the highway, so that i didn’t pay inflated near-campus prices. that was probably a bad call, given the trends of the day, but i did manage to pull that off successfully.
seriously, i’m feeling very un-awesome right now.
i’m spending this uber-exciting night putting together lots of pretty data on a poster for a rapidly-upcoming conference, while guzzling caffeine to ward off the headache and antihistamine to ward off the awful lab rat allergies… and i’m tagged on some blog game thingy by Abel Pharmboy.
(damn. i can’t not capitalize everything on the blog, can i?)
so here’s what i’m supposed to do:
1. Link to the person who tagged you.
2. Post the rules on your blog.
3. Write six random things about yourself.
4. Tag six people at the end of your post and link to them.
5. Let each person know they’ve been tagged and leave a comment on their blog.
6. Let the tagger know when your entry is up.
so. let the six randoms fly…
1. in college, i used to bike 100+ miles per week. i only drove my car to my retail job in the next town or to the grocery store once a week- usually on my way home from work. now i drive entirely too much because i’m afraid to bike here. i have checked out some nice local trails, and would definitely hit those up often… except that when it’s not too hot outside, it’s too dark when i get home. i just can’t win.
2. i haven’t gotten a speeding ticket since i was 17, only because i have learned to keep an eye out for the cops- not because i quit speeding.
3. i am such a tomboy- always have been. i like to get my hands dirty, can handle uncooperative large animals, like to take things apart, can school you on how your car works, and throw a hell of a punch.
4. at a big name private fancy-pants university like the one i attend, i’m a total misfit. i’m okay with that, because i’m just using the experience for my own betterment. i don’t really care what people think of me on a personal level, which makes it easy to mess with my fellow students. [evil laugh]
5. three good friends of mine are serving in afghanistan. one of my friends from college is currently somewhere in europe with the usaf. one of these good people is my self-adopted big brother. every day i think of them and hope they’re doing ok. i don’t have much money, but i send goodies from home when i can. you can send a flat rate “large” box to afghanistan for about $10 in postage.
6. my favorite book is catch-22 by joseph heller. if for some reason you haven’t read this classic yet, i highly recommend it.
is it bad luck to break the tag chain? i don’t know enough people well enough to tag six of them.
best voting experience ever: absentee ballot.
no standing in line. no asshole [other candidate] supporters screaming racial epithets and slashing tires. no going way the hell out of my way, no missing time at work. fill in the circle, drop in the mail.
hell yes.
also, beats driving 2000 miles round trip to the polling place in another state.
let’s talk about the amount of work it takes to make raw data mean something. i just spent a lot of time on that myself.
i get some numbers that represent radioactive counts from the scintillation counter, that correlate to whatever i put in the vial that was counted. but if i were to just hand that long list (literally, about 40 feet of data) over to someone else, they’d hand it back and ask what the hell all that meant. so the idea is to put it all into meaningful terms in a graphic: something that you can look at, analyze, and hopefully points out the super cool discovery you just found with those hundreds of hours of work.
i’ve talked before about the lovely sigmoidal dose-response curve that is a hallmark of pharmacology. those don’t come about from going through your excel files and dumping values a-x into graphing program cells 1-24. you organize the data properly, transform your drug concentration to a log scale, and that one point where you maybe failed to add the radioactive tracer gets excluded for having a count of 12.2 cpm (on average i’m seeing in the 15000 cpm range for my higher counts) and then you get to fit a curve using the graph program. you have now taken raw data and expressed it as, say, a specific cellular response to a certain concentration of a drug. progress! now it means something more than numbers on a sheet of paper.
ok, so now you have 80 dose-response curves that constitute your study. awesome. shall we send it to the journal of neuroscience, then? of course not, they’ll tell you the same thing you heard when you handed the big list of numbers to someone. wtf does this mean, have it back. now you dig deeper, analyze further. put groups together, get stats, etc.
so then you play with adding these curves together while maintaining statistical integrity- personally, i add each curve as its own replicate in a set. some may calculate mean, sem and n and input them separately- which is equivalent, with the particular program i use. but either way, comparing even 16 curves is cumbersome and time-consuming for someone who just wants to know what your results say in an easy-to-understand format. you don’t want someone’s eyes to gloss over when they look at your data. so you need something that combines all this information into one “pow!” kind of figure.
that’s right, hundreds of hours went into collecting this data between injecting mass groups of rats and assay calibration, completion and data input… and i want to make it one graph. one figure on the poster. the upside is that everyone who sees it will indeed know how much time it took me to produce this one graph- that’s the great thing about science.
so tonight i tinkered around with the stuff that was different across experiments- this was the point i wanted to highlight. and i got rid of the dose-response curves altogether. instead, i focused on one area: maximal effect of drug under differing conditions. since we always have controls, i expressed this as a percent of control effect and compared all groups directly to one another. pow, the data is now directly comparable across the board. and pow, that difference pokes you in the eyeball when you look at the figure.
and that is exactly what you want. more is not always better, though i still need to continue reminding myself of that sometimes.
now i’m on to the next experiment (though i have some more tinkering with this data to do, along with another big analysis- nothing like a grant or progress report deadline to dredge up projects you thought you had abandoned for lack of interest!) and i may be light on the blogging for awhile as this new stuff and my rat allergies once again take over my life.
i’ve been dealing with this sustained overworking-myself stress and putting myself up to analyzing all that data in large chunks- then give myself a few days to recover during which the nausea starts to subside. then the boss sends one email. one email! and i think i’m going to throw up again. it wasn’t even something that should stress me out- but there is that feeling again.
something is wrong with me for willingly doing this to myself.
i have a paper to read before i go to sleep tonight. while i try to not throw up. if that’s not gluttony for punishment…
life here at big name u can be very frustrating at times- you know it’s good for you to have the bar set very high, but sometimes with the accompanying stress, you want to take that bar down and smash windows with it. the people around you are the best and brightest and sometimes it’s easy to feel like a complete and total idiot in comparison- especially if you’re having a bad day (week, month, etc.)
but there are some major upsides to life here, like having some really awesome people come to give talks and workshops. i have attended several career and personal development workshops given by high-profile people in the past couple of years, and i am anxiously looking forward to the one coming up next week by a columnist in a major publication that i have been reading closely for over a year.
if there was anyone i ever wanted to talk to about transitioning from academia to industry, this would be the person. and next week i have the chance to do just that. how unbelievably cool.
i have two massive complete data sets, now analyzed, printed, and sitting on my desk. i feel vastly more relaxed than i have felt over the past few weeks. maybe the past few months. for a brief moment in time, there are not a handful of things pulling my attention in opposing directions. this is a very brief moment, but enough time to regroup.
the boss’s reaction was exactly what i wanted to hear. i was confident in my own work (or what’s the point of doing it) but hearing that outside vote of confidence is always nice.
the List has one fewer item left to do. this is the best part of all. second best is zeroing out my radioactivity logs, i loathe all that paperwork.
with any luck, today was the last day of absolute and total madness related to this particular project. on top of the day’s usual insanity, i also decided to get going on getting all that data into excel files. manually entering 72 curves consisting of 48 numbers a piece is a serious bitch. i got all stubborn and refused to leave until it was finished. sometimes i really hate myself. it took me several hours on top of the usual day’s work, even after my undergrad assistant spent a couple of hours on it too.
i stumbled, feeling nearly braindead, out of the lab and to my car; realizing that it was fairly late and i hadn’t eaten since noon. fortunately, one way to get to my cozy little apartment from the university happens to take me right past a little caesar’s full of hot ‘n’ ready pizzas for cheap. i took home a pepperoni pizza and mowed down half of it. i swear that’s small for a “large”- or i’m just used to the new york pizza place across from campus with the massive pizzas.
that’s probably my only frivolous expenditure for the week.
i had considered taking some time to put the data into my copy of the lab’s preferred data analysis software, but looking at how difficult this was to type… i think i’ll do that another day. i need rest.
the actual number of ways to screw up your experiment is infinite. but as i went through the daily grind and also as i discussed a fellow student’s work today, i thought of this list.
1. failing to plan ahead, or planning for things to arrive when they’re supposed to.
you really have to have all your ducks in a row before you even think about running the experiment. have everything ready, all solutions made, staring you in the face on your bench. if you ordered new chemical yesterday, do not assume it will be here today even if you asked for overnight shipping. it will probably not be, that’s murphy’s law and you become intimately familiar with it as a graduate student.
2. cutting corners
if you don’t have time to do it right the first time, how do you find time to do it again? suck it up and do things properly the first time, even if it takes more time and effort. also, it’s stressful when shit doesn’t work. that causes you to screw up other things. get it over with by doing it right and not skipping steps.
3. not paying attention
binding assays are boring as hell, i’ll say it a thousand times. i can crank em out as needed, but even bored out of my mind i have to pay close attention. otherwise i could forget to add something, screw up a dilution, and throw out my entire day’s work (not to mention the sample tissue which is a pain in the ass to collect.) i have a way of keeping track of what steps i have completed, but attention is key.
4. not using a timer
there is a difference between an hour long incubation and an hour and twenty minutes long incubation. if you can’t keep track of time (this is especially common during lunch hour, when conversations get started) then grab a timer. also, in the lab environment, a beeping timer is a socially acceptable reason to leave the conversation immediately.
5. math!
a simple math error is easy to overlook when you just spent a lot of time poring over your equations, inputs and outputs. do the math twice, separately. if the answers match, you’re set. if not, time to start over or consult a colleague. i have never taken issue with someone politely asking for a minute of my time to review a set of calculations- with a bit of advance notice.
6. skipping lunch
i’m guilty of this myself. and the hungrier i get, the worse my pipetting gets. unfortunately for me, i’m usually trying to set up a protein assay around lunchtime. ideally, it’s far more efficient for me to eat during the development period. but some days, i just go eat first because i can’t pipet a good triplicate with shaky hands. it took me ages to figure this out, sadly!
7. overdoing the multitasking
sometimes it seems like maybe you’ve got just a little bit too much downtime in the middle of your day- why not find something else to do? juggling multiple experiments is a great way to get more done with your time- sometimes. but if you try to do too much, you might have a little too much up in the air and drop the ball. which brings me to…
8. not knowing your limits
sometimes it seems like you could just force yourself to keep working all day and all night, and crank out the data like a machine. i’m guilty of it myself, but i know my limits. when i was a younger grad student, i didn’t know them. and i wasted a lot of time, energy, supplies, research funds trying to do what i was not capable of doing. i also got pretty frustrated.
9. hasty experimental design
you know what you want to find out, but is your experiment designed to tell you that answer? do you have proper controls and relevant doses? take some time to review your experimental plans- this is a great discussion to have with your principal investigator. mine loves to talk experimental design, especially if there’s a developed one presented for critique. this is also a common subject in lab meeting.
10. forgetting about small nuances
everything you do has some small nuance associated with it. a particularly good example is the behavioral study- rat behavior is very sensitive to changes in time of day, order of daily routine, experimenter handling, everything. for this type of experiment, wearing a different color lab coat and grabbing a slightly different gauge needle might seem insignificant but may be enough to affect your results. you should set up your work environment to help you do things correctly, which ties back into planning ahead.
i know there are about a billion more ways to screw up an experiment, and these are very general. i can list off about 100 ways i have screwed up western blots alone! mistakes happen. the important thing is to learn from them- that is what grad school is really about.
Recent Comments