World-Wide
Day in Science:
A Snapshot from Edinburgh, Scotland
On April 15th, 2004, two postgraduate research students from Edinburgh
University, Scotland, kept diaries of their day in science.
Michelle Galea is from the School of Informatics and in her second
year of a PhD in Artificial Intelligence. Natasha Tian is in the
first year of a PhD in Developmental Neurobiology looking at genes
that help form the developing brain.
Here's how they spent the day.....
Discovering
the World of Artificial Ants: A Day with Michelle Galea
9:10am
Today is Thursday 15th April 2004, Worldwide Day of Science, and
I've been asked to write about how I – as a scientist –
am going to spend this day.
First, a little about myself. My name is Michelle, I'm female, 36
years old, and after years of working in 'the real world' –
as non-academics like to call it – I've gone back to university.
I'm a postgraduate research student in the School of Informatics,
University of Edinburgh, and in the second year my PhD in Artificial
Intelligence. I thought I would be the oldest student here and was
amazed to find there were several others even older. The diverse
nationalities is another great thing...I share an office with 7
other students and our nationalities are Maltese (that's me), Irish,
Malaysian, English, Scottish, Greek and Chinese. And that's just
in my office.
OK. So these are the things I need to do today:
1.revise paper accepted at Budapest conference. Wonder what real
ghoulash will taste like when I get there...mmm....
2.get my artificial ants to work harder, faster, better! More about
those clever little things later...
3.reply to the grad student from Texas who emailed asking for more
information about my work. Must admit it's nice to know other people
out there think my work is both interesting and useful....
But first, a cup of tea I think…
9:35am
I'm off to my first big conference in Budapest in July, but before
that I need to finish the necessary revisions. The revisions are
a result of the suggestions made by external experts in the field
who reviewed my work. The reviewers perform a quality control task,
ensuring new work is up to an acceptable standard. I've managed
to incorprate their very useful comments without too much difficulty.
The annoying thing is getting everything to fit on just 6 pages!
I need to reword some paragraphs as it's now overflowing onto the
7th page by a few lines.
10:10am
lines removed: 8 – v.good
lines added: 12 – v.bad
... 10:50am
lines removed: 16 – excellent!
lines added: 19 – how on earth....?
Must remove 3 lines, but first a cup of coffee I think....
12:00pm
Somehow managed to squeeze it all onto 6 pages. It's amazing how
much time scientists seem to spend trying to adhere to these page
restrictions! But, I suppose it does mean there's less opportunity
to ramble on and on about very little. I'll read the paper one more
time, print it one more time and give it to my supervisor for one
more check.
12:20pm
Bumped into Laura while I was picking up my printout. She's another
2nd year PhD student here. The School has been interviewing people
for a lectureship position. Today, as part of the selection process,
the interviewees need to give a presentation about their work. Laura
asked whether I knew where the presentations were being held. I
didn't. But yesterday one of the senior researchers suggested this
might be interesting to us so we decided to find out. We've missed
2 presentations but there are going to be another 2 today. The next
one starts at 12:30pm – only a minute's walk away, so Laura
and I are off to see what this sort of thing is like. We'll need
to find a job after our PhDs and one day we might be making these
presentations ourselves.
2:00pm
Had a working lunch at my desk (delicious moussaka and salad from
a diner closeby), and looked up some recent journal papers that
seemed to be relevant to my work. Have 20mins to start setting some
ants working before I'm off with Laura for another presentation.
Speaking of which...
The presentation was interesting, not because we understood the
woman's area of research in detail, but because of the circumstances.
There were about 20-25 people in all, mostly senior researchers
but a few graduate students like Laura and myself. The interviewee
was a young woman who seemed a little nervous at first (who wouldn't?)
but managed to talk about her work quite easily. Laura and I were
impressed by how she handled the sometimes-difficult questions,
and agreed we would have found the experience an ordeal. I suppose,
though, it's not that different from other job interviews you might
go for, and preparation is key.
3:35pm
Surprising how much you can do in 20mins when you think you don't
have any more time to waste. Managed to set up several experiments
before I left and now I can start finding out what those ants have
been up to. I'll tell you a little about my work first, though…Right
after a cup of tea…
My broad area of research is AI – Artificial Intelligence
(a vet friend of mine recently pointed out that I shouldn't just
say 'AI '. To her, AI means Artificial Insemination so I'm more
careful nowadays!). Many AI researchers have been concerned with
understanding human intelligence and replicating that in machines.
But now, some of us have shifted to mimicking social insect behaviour.
I know this may seem odd, but think about it for a while –
as intelligent as humans are, there are some large-scale jobs at
which insects like ants and termites are much better.
Take a real ant colony for instance, numbering thousands or even
millions of ants with apparently no one individual in charge. And
yet simple actions by these individuals lead to complex and life-saving
decisions for the colony, such as finding food. There are usually
many ants trying to find food, some get lost, and a major form of
communication between is indirect (they lay a chemical trail behind
them so that other ants may follow). Yet, almost blind and travelling
over great distances, the colony is generally capable of finding
the shortest path between its nest and food.
Now my problem is not foraging for food, but making sense out of
mountains of data. This could be data generated by credit cards,
for instance, or observations of physical symptoms of patients.
So, I set loose an army of virtual ants on this electronic data,
with the task of foraging for indicators of fraud on the one hand,
or breast cancer on the other. The results we're getting are very
encouraging, but still a lot of work to do.
More Antificial Intelligence than Artificial Intelligence I suppose.
4:30pm
Aagghgh! Qiang (my supervisor) walked in with what he says are some
very minor revisions to the paper. Word and line counting again!
4:35pm
Thankfully, his few extra words did not mean an extra page.
6:25pm
Well, my ants are doing very well on one problem, but not so well
on another. I'll need to devote most of tomorrow trying to find
out why (no flitting off to hear presentations!). I'll also need
to come in earlier than usual to prepare for my 9:30pm meeting with
Qiang. We'll discuss what I've been doing, how well it's been going,
possible ways of resolving problems, and what goals to set for the
next few weeks.
And, I've just emailed Xuepeng. He's a Masters student in the States
and came across my own Masters project using artificial ant-like
software agents for knowledge discovery. He'd like to do something
similar and has asked for advice (when i was doing my own Masters
project I was in regular contact with a Brazilian scientist on whose
work i was building). Xuepeng would also like to use my software.
I can suggest a few things (including also getting in touch with
Ronan, one of the other PhD students in my office who worked on
a similar subject area last year), but I'll need a few days to dig
out my old code and make sure it's understandable for him to use.
This is one of the things I find most enjoyable about this work,
interacting with people from around the world. I thought that life
as a PhD student might be a solitary experience (and it can be,
if you want it to), especially when you start drilling deeper and
deeper into your own particular problem area. But, really, email
and the internet has made communication with others working in related
areas so easy.
It's almost 6:30pm and I'd better pack up. Hubby's picking me up
on the way home in a little while. I think it's his turn to cook
tonight. I can play with the pussy cats instead.
Growing Brains and Other Nerve-wracking Stuff:
A Day with Natasha Tian
The sweeping
melody of Vaughan Williams’ ‘The Lark Ascending’
permeates my semi-consciousness. Opening my eyes as little as possible
I glimpse the time: 7.05 am. “15 more minutes”, I tell
myself, relishing a precious 10 whole extra minutes underneath my
warm duvet. By 8.30 I am ready to leave in my daily uniform of blue
jeans, comfy trainers and whatever-top-happened-to-be-clean.
Walking across
the Meadows is wonderful at this time of year – wafts of spring
air reach my nose, blossoms cover the trees lining Middle-Meadow
Walk and the sky is blue. A hot blast of air hits me as I enter
the Medical Buildings, which never fail to be over-heated. I reach
the Developmental Neurobiology Laboratory and punch in the door-code.
When I first started my PhD, I was frustrated to find that each
different part of the lab (histology, tissue culture and the microscope
room, to name but a few) had its own access code. I suspected that
memorizing these door-codes was the first test for all new PhD students
and I feared that secret cameras were constantly recording my futile
attempts to gain entry.
My lab is fairly
large. We share lab space with three other groups that work on projects
with a common developmental theme. This means that there are always
plenty of people to ask for advice. My laboratory works on genes
called transcription factors that play a role in forming the developing
brain. I am 8 months into my PhD, investigating the effects of a
gene called Foxg1 on the guidance of retinal ganglion cell axons.
This is not the area I expected to find myself in but after a short
Master’s project in this lab, I knew I had to return. The
buzzing lab atmosphere, excellent supervision and of course the
project itself are second-to-none.
After the obligatory
e-mail check and cup of tea, I consult my diary; without which I
am completely lost. Although it tells me what I should be doing,
in reality my day often follows a different course. Initially ‘short’
or ‘simple’ experiments turn out to be longer and more
complex than predicted. The words ‘photoconversion’
and ‘check cultures’ are pencilled in for today. My
first priority is to book all the pieces of equipment I need: the
vibratome and tissue culture hood.
The word ‘vibratome’
conjures up images of an anti-cellulite treatment or one of those
devices that gives you small electric shocks to tone you up while
you ‘relax’. In reality, it is a machine that cuts up
pieces of tissue (small brains in my case), using a vibrating blade.
It takes me a long two hours to cut three embryonic mouse brains
into 200 micron-thick slices, as the blade moves forward very, v-e-r-y
slowly. Next, I need to transfer the brain slices through various
solutions to prepare them for photoconversion. This process will
convert the fluorescent dye that I used to label neurons from the
eye to the brain, into an insoluble precipitate using strong fluorescent
light. My stomach gremlin gurgles suddenly. Luckily I can escape
for lunch whilst my brain slices are equilibriating in ‘solution
1’.
If you step
out of the main entrance to the Medical School, there are a zillion
different places to eat. My favourite has to be ‘Rudi’s’
with its all-Italian sandwiches and the cute guy serving who changes
his hair-style like a chameleon. Today he sports a black mohican
and I am so distracted that I fail to order my sandwich coherently.
Finally I manage to point and say, “That one please”,
feeling my face reddening. After lunch and the usual office banter,
which tends to involve planning our next lab night out, I decide
to check my retinal cultures from three days ago. My aim is to coax
special cells in the retina, called retinal ganglion cells, to project
nerve axons. Since the pieces of retina contain a fluorescent tag,
I need to examine them under the fluorescent microscope. However,
the room needs to be pitch black in order for me to visualize the
nerve axons that I hope to see. As I make my way in the dark back
to the microscope, I inevitably walk into a chair. This happens
with such regularity that I can now recall the date of each culture
experiment simply by examining the colour of my bruises. Surely
that alone is worthy of a chapter in my PhD thesis. To my astonishment,
I can see faint nerve axons sprawling outwards from the retina.
Distressingly, they seem to be floating around like a jellyfish
rather than sticking to the bottom of the culture dish. I carefully
walk towards the incubator imagining that I have a book on my head
to cause minimal disruption to my delicate axons. I rush to tell
the supervising postdoc of my exciting findings, only to discover
that after moving my cultures back and forth between the microscope
and incubator, most of the axons have been sheared away from the
retina. As always, the postdoc is sympathetic about my failed culture
attempts and his optimistic suggestions give me hope and determination
that the next experiment will be a success.
Scientific research
appears to advance at a snail’s pace, which makes life in
the lab frustrating at times. Nevertheless, I am able to control
the direction of my own research project and constantly revise or
create hypotheses in the light of new discoveries, making life as
a scientist totally absorbing, highly dynamic, and rarely dull.
Of course I have my fair share of bad days when nothing seems to
work, testing my patience to the limit. But given the choice between
science and a different career, I would not trade in my white lab
coat.
