Death Stare for Gullie

It’s happened to me. I’m sitting, calmly enjoying a sandwich outside on a bench, and then …

A seagull swoops in and tries to steal my food.

It’s terrifying. Seagulls are scary, especially up close and especially in Dundee, where I used to live and frequently sit outside eating something or the other. They have mean eyes.

I remember the seagulls in Dundee being quite peculiar. An anecdote: I was walking along the sidewalk, edging close to a corner where a seagull was digging through a ripped trash bag. When I was a few meters away, the seagull looked up and did this little walk away from the bag, pretending as if they weren’t just digging through trash. After I passed the corner, I glanced back and saw that they’d done a u-turn and went back to digging.

Okay, maybe I’m giving the bird too much of a personality. But it was weird.

Back to the food stealing; a research conducted at the University of Exeter showed that if you stare at a gull, it is less likely to steal your chips (for US readers: french fries which are totally not from France but from Belgium and stop calling things the wrong name and, never mind, I’m okay).

Granted, the study had a limited scope. They tried to test 74 gulls, but more than half of them flew away. And it is likely that a lot of seagull related crime is due to a few bad seeds and most seagulls are perfectly happy leaving you and your food alone and digging through trash for snacks.

Nevertheless, seagulls that were “looked at” while they were approaching food, were a lot less likely to touch that food. In fact, only a quarter of seagulls that were being watched while they tried to approach and eat food actually touched the food.

Maybe they were just scared of getting caught while committing food theft. Maybe they hate the color of our eyes. Maybe our stare is truly terrifying (I certainly know a few people with a scary stare). But next time you see a seagull approaching your food, give them the death stare. Perhaps your meal will be saved.

Come at me bro



A few weeks ago, I went to a public lecture – attractively named “Wine Down with Science” – organized by UW Medicine and I’ve been trying to tell people about it ever since.

The first problem is that the topic of the talk is one of those words that I’ll always just struggle to pronounce on the first go:


I went to the event not knowing what it was going to be about. I was already sold when I read it was a public lecture about some ground breaking research; any excuse to listen to science in a more informal setting. Turned out, it was good of me to go, because it was sort of about biophysics and biophysics is sort of my jam.

So what is the *insert long word here again*?

Proprioception you mean? According to the event announcement, the lecture title was: Out of Your Mind: The Inner Workings of Your Mysterious Sixth Sense. So proprioception is about seeing dead people?

Sorry to break it to you kid, but that’s not really a thing.

Nope. In reality, proprioception is our ability to be aware the position and movement of our body and its parts. Not having proprioceptive abilities is one of the reasons that man-mimicking robots fail at a lot of seemingly basic tasks. They have no internal feedback system to tell them how their parts are positioned or moving with respect to each other, making benign tasks hard to do.

I feel bad for laughing but it is very funny.

With proprioception, we know exactly how are body is positioned: whether our arms are bent, our feet are flexed, our eyelids are closed (okay, there are other ways to tell). And that without having to look those body parts. There are some known cases of people losing their sense of proprioception and it causes paralysis (if you have 10 minutes to spare, you should really check out this video about a man who lost his sense of proprioception but taught himself to walk using visual cues instead).

How can we study this proprioception thing?

This is where it gets even cooler. The lecturer – John Tuthill – explained how in his lab, they use a clever combination of lasers, genetic tools, virtual reality, and fruit-fly-treadmills to understand how proprioception works.

Fruit fly on a floating ball tread mill. I can look at this for hours. From:

By making a fruit fly run on a floating ball, and surrounding it with screens, they can trick the fly into thinking it’s strolling somewhere outside and track the neuron activity during movement. Using a laser, they can turn off the proprioceptive neurons very locally (using something called optogenetics, but that’s for some other time). For example, if by blocking proprioception in one of the fly’s legs, i.e. stopping communication between the leg and the central nervous system, they temporarily paralyze that leg. After turning the laser back off, the fly trods on as if nothing has happened.

Why should I worry, why should I care?

(Any excuse to play that song).

First of all, part of science is just understanding how things work. Knowledge for knowledge’s sake. But there are also some useful applications of this knowledge, such as helping people with movement disorders. And helping robot-designers to not make robots that seemingly stupidly fall over.

On the other hand, we can make robots do Walking manalready. Source: Boston Dynamics

This talk was by John Tuthill, PhD, Assistant Professor of Physiology and Biophysics. His work is really cool. Go check it out.


You might have noticed how I’m a low-key (*ahem*) LEGO fan, but only if you’ve really been paying attention, and it has been quite a while since I went on a Lego-nerd rant (hm, not really).

Anyway, this 86-year-old company, named for the Danish words for “play well” (“leg” and “godt”), is not only known for its iconic building block but also for its iconic minifigure:

Decomposed LEGO figure. Image from (1)

Unlike the building block, which has remained the same for decades (in fact, a current block will still click with a 60-year-old one), the minifigure has gone through some major changes. Between 1975 and 2010, there have been at least 3655 different minifigures, and in 2000 there was an estimated total of over four billion mini Lego people! Actually, according to some predictions, the total number of minifigures will surpass that of humans next week!

And I trust Randall Munroe’s math… Image from (2)

The first minifigure (1975) didn’t even have any moving parts. It wasn’t until three years later, when the familiar yellow smiley-faced figure came out (your friendly neighborhood cop), that the arms and legs could be moved:

1978 guy (right) says: “Hands up!”
1975 guy (left) says nothing, for he has no face. Image from (3)

This was also when the Lego minifigure hand shape was developed. It is very useful for holding Lego-things. However, is very inefficient when your trying to be sarcastic…

Image result for quotation hands lego gif

Throughout the years, the figures have gotten increasingly more complex. It started with the hair (early male minifigures wore hats until the hairpiece was designed in 1979), then the accessories arrived. Hats, bags, hand-held weapons, … the whole shebang. In the meantime, the outfits got more detailed. When licensed playsets started taking off near the turn of the millennium (the first Star Wars series hit the shelves in 1999), a whole franchise originated which would include books, video games, and animation films. More series soon followed, including Harry Potter, Batman, LOTR, … the list is endless. And while the outfits and accessories became more elaborate, the faces became more – well – emotional?

Initially, there was only the blank smile. Now, Lego minifigure faces encompass all the emotions. According to a 2013 paper, there are six main types of facial expressions: disdain, confidence, concern, fear, happiness, and anger.

Scorny, Gutsy, Worry, Scardy, Happy, and Grumpy. Dopey has gone missing. Image from (1)

Happy and angry faces are the most common, with the relative proportion of happy faces decreasing over time. In short, with an increasing emotional range, Lego minifigures seem to be getting more “human.” Soon, they’ll have Myers–Briggs personality types!

Over But for now, we’ll have to do with two-faced police officers…

Over time, Hollywood movies have gotten increasingly less black and white (I do not mean the colors), with multi-dimensional characters, heroes with a dark side or villains that seem relatable, and it seems that the Lego minifigures are following suit.

In any case, Lego might be over 80 years old, and the minifigures over 40, but no matter how old you are, you are never too old to play, build or tinker with Lego!

Adult Valerie admiring the Lego store window. (Photo by Lale)



(1) Bartneck, C., Obaid, M., & Zawieska, K. (2013). Agents with faces – What can we learn from LEGO Minfigures. Proceedings of the 1st International Conference on Human-Agent Interaction, Sapporo pp. III-2-1.

(2) xkcd


What is this “science” thing?

I’ve felt bad all week. Well, not really all week. And not really bad. I’ve felt a teeny bit guilty for joking that economic sciences is not really a “science”. The “soft” sciences (social sciences, economic sciences, psychology, to name a few) are too often ridiculed by practitioners of the “harder” sciences. I’ve done it too. Last week in fact, as I’ve just said.

If the sciences were ranked (by xkcd)

Most of the “soft” scientists I know don’t really mind too much (yes, I have soft science friends, I *can’t* be an elitist), and they laugh along. But still, I wanted to bring a bit of nuance and perhaps a tiny apology,


especially since this years’ Nobel Prize for Economic Science was awarded for integrating climate change an technological innovations into long-run macroeconomic analysis. Two subjects that are kind-of STEM-related.

Therefore, no matter how you might be willing to rank them, something can be considered science (from the Latin word scientia – “knowledge”) if the scientific method is applied.

What’s this scientific method?

The scientific method is a way to approach a problem or question by following this – or any similar – flowchart: *

Example of a scientific method flowchart

Very briefly and with an example, these are the steps you’d follow:

  1. Observation
    This can be anything you observe.
    Example: People seem a lot friendlier here in [town A]. When I pass people on the street, people smile at me more than they did when I was in [town B].** 
  2. Question
    From that observation, you can formulate a well-defined question, a problem you would like to know the answer to. Science is simply the pursuit of knowledge, you know.
    Example: Are people more friendly in [town A] than in [town B]? (if friendly is defined as “smiling at people on the street”) 
  3. Hypothesis
    You probably have a little bit of data (from your observations) that allow you to formulate the answer you would expect. This possible answer is something you can test: is what you assumed true or false?
    Example: People in [town A] smile more on to passers-by than in [town B] 
  4. Experiment
    Now it is time to collect your data.
    Example: I’d go to [town A], walk around in the center for – say – 30 minutes and count how many people I pass on the street (and actually make eye contact with) and how many people smiled at me. I’d then do the same for [town B].

  5. Analysis
    When you have collected all your data, sit down and perform some analysis. Usually, statistics are the thing to apply.
    Example: I’d calculate the ratio of smiling people in each town, let’s say 17 out of 59 (29%) of people smiled at me in [town A], while 34 out of 81 (42%) people smiled in [town B].

  6. Conclusion
    Example: I reject my hypothesis; people in [town A] are not friendlier than people in [town B].
    This last step is checking if my hypothesis was correct (it wasn’t). Rejecting the hypothesis means I can go back and change my hypothesis and start again. If my hypothesis was correct, yay – I’ve done science!

Well, in reality, there is even more to it (both for rejecting and accepting an hypothesis).

In this example, there are many faults. Was my definition of “friendliness” correct? Were there factors I didn’t account for, like a bit of spinach between my teeth that caused more people to smile (or laugh) at me? More importantly, if I repeat the experiment, do I get the same result***? Was my experiment well designed; maybe there are better ways to test this same hypothesis?

Back and forth and back and forth and back and forth again.

Science is a very iterative process. Hypotheses are constantly being reformulated and retested. It is actually impossible to be 100% a hypothesis is true. The real science is when you try every which way to disprove your hypothesis. It is after a lot of back and forth and iteration, that a theory about something can be formulated. But you should know that in the scientific lingo, a theory has nothing to do with guesswork. It is the result of several repeats of observations and experiments that are generally accepted as reliable accounts of the world around us. ****

Scientist vs. engineer

I’d also like to note that science and engineering are quite different things. A scientist wants to know how things work while an engineer kind of just wants to make things work.

For example: engineers built the large hadron collider; scientists use it to study elementary particles.

Though it should be said that a lot of scientists have a bit of engineering in them, and vice versa.


* Listen to Niel Degrasse Tyson explain the scientific method using a whipped analogy below.

** I just know that this is just because I look funny.

*** Typically, at least three repeats showing the same conclusion are necessary to accept a hypothesis.

**** More about scientific theory here:


An ignobel cause

Disclaimer: if you’re a bit hungry and/or know that reading about spaghetti will make you hungry, I suggest you go eat some spaghetti before you continue reading… But if you do, keep at least a few strands uncooked, you might need it later on.

An odd article popped up on my go-to news site the other day. And then the day after that, an article on the same topic popped up in the newspaper I was reading. It was an article reporting on the science of breaking an uncooked spaghetti.

No, I’m not joking.

And apparently, the research solves a decade-old problem. I never knew spaghetti could pose a decade-old problem, except for maybe the secret spaghetti-sauce recipe of an Italian-American family but that’s a century-old problem, I would say.

So if you’d go into your kitchen now, take a strand of uncooked spaghetti, hold it at the ends, and start bending it until it snaps, you will see what this mystery is all about. Most probably, you have now ended up with three or more bits of spaghetti. If you are super bored or think snapping spaghetti is super-fun (this is what Richard Feynman apparently thought), you can try it again. And you will notice the spaghetti almost never snaps into two pieces. Or you can just take my word for it…

In 2005, some French physicists came up with a theoretical solution to why spaghetti never breaks into two, because this unsolved mystery Richard Feynman broke his head about merited some further research…

When a very thin bar (or strand of spaghetti) is being bent, this will cause the strand to break somewhere near the middle. This first break will cause a “snap-back” effect which essentially causes a vibration to travel through the rest of the strand, causing even more points of fracture, which results in three or more pieces. In other words, is very rare to end up with exactly two pieces of spaghetti.

These French researchers were rewarded with an Ig Nobel prize for their finding. An Ig Nobel prize is a prize that is rewarded “for achievements that first make people LAUGH then make them THINK” and also the reason for my best quiz achievement ever.*

Experiments (above) and simulations (below) show how dry spaghetti can be broken into two or more fragments, by twisting and bending. (Image: MIT)

And now, years later, mathematicians from MIT have added to that research by coming up with a way to ensure a dry spaghetti strand does break exactly in two: by first twisting the spaghetti before bending it. The twisting part causes stresses in the spaghetti strand that counteract the snapback effect when it eventually breaks. When the spaghetti does break in to, the energy release from a “twist wave” (where the spaghetti pieces untwist themselves) ensures there is no extra stress that would cause more fracture points. So there we go: the spaghetti breaks in exactly two pieces as long as you twist it enough.

Experiments (above) and simulations (below) show how dry spaghetti can be broken into two or more fragments, by twisting and bending. (Image: MIT)

Now, this theory isn’t only limited to breaking spaghetti. Understanding stress distributions and breaking cascade also have some practical applications, according to the authors: the same principles can be applied to other thin bar-like structures, such as multifibers, nanotubes, and microtubules.

Now, if you haven’t already, go get yourself some spaghetti.


* The question: who has one both an Ig Nobel and a Nobel prize and for what?
The whole table looked very confused and I just said very confidently “André Geim, levitating a frog and graphene” so it turns out a degree in nanotech is super useful for winning quizzes. (Actually, I’m not even sure we won and I doubt it was thanks to me answering that one question correctly, but I’m pretty sure I will never live up to that moment ever again.)

The devil’s in the details

One of the “hallmarks” of cancer is the ability of cancer cells to spread to other parts of the body, settle themselves in this new environment and give rise to a new tumor. This process of spreading is known as metastasis and is something that typically aggressive cancers are known to do. In almost all cancers, cells can only spread within one organism. Almost all. There are a few – so far 4 that we know of – types of cancer that can spread to another body. In other words, there are types of cancer that are contagious, or transmissible, and that’s kind of creepy.

A transmissible cancer is a cancer where the cells themselves can spread to another organism and cause tumor growth in that organism. This is not the case for virus-born cancer. For example, in the Human Pappiloma Virus (HPV) is a virus that can be sexually transmitted and some types of the virus can cause a whole range of different cancers.  In other words, the virus is transmitted and the virus gives rise to cancer development.

But in the case of transmissible cancers, it is the cancer cells that spread to another organism. Most types of transmissible cancers are sexually transmitted; these types are found in snails and dogs. There is one type of transmissible cancer that is a bit of the odd one out: devil facial tumor disease or DFTD. Sounds kind of satanic, no?

DFTD is a very aggressive non-viral transmissible cancer that affects Tasmanian devils, you know, that Looney Tunes character that creates little tornadoes when it moves…

Why you so aggressive, Taz?

Okay, not really.

DFTD is a mouth cancer that looks really bad (don’t google it, or do, whatever, I’m not your boss) and is spread because Taz devils bite each other a lot. I guess, it kind of is an STI because they also bite each other during mating.

And because Taz devils are pretty isolated (they all live on the one island), and the cancer is very aggressive (spreads easily and is very lethal), and Taz devils are pretty aggressive animals (they bite each other a lot), it is a bit of a problem. DFTD has been observed since 1996 and has now spread to most parts of the island. It is feared that DFTD may cause the extinction of Taz devils.

Last year, I went to a talk by Elizabeth Murchison, who studies transmissible cancers, and it turns out that DFTD is actually quite interesting. In her talk, she explained that her team used genome reconstruction to track the origin and evolution of DFTD, and this led to the discovery that there are two independent types of DFTD (if I remember correctly, one of the cancers originated in a female Tax devil while the other originated in a male and that’s how they knew it had to be two separate types of DFTD).

Why would I care? Well, first of all, it is a unique situation to have a transmissible cancer that is isolated to one island, which – scientifically – is a unique opportunity to study how cancer evolves and spreads. Moreover, it is pretty strange that there are two types of a rare cancer (transmissible cancers are very uncommon) that have originated within one species. The two types of cancer started in similar tissue types, and have similar mutation patterns, which implies that Taz devils may be particularly susceptible to transmissible cancers.

But then the question is: why now? Taz devils have been around for ages, why have they now, within what seems to be only decades, developed two different diseases that are very similar to each other and that may lead to their extinction? Is it due to human influence, or perhaps climate change*? Has this happened to other species before, but we just didn’t know because we weren’t around or we didn’t know about cancers yet?

And, can we save the Taz devil? There are efforts to set up Taz devil sanctuaries on smaller islands off the coast of Tasmania to avoid these cancers spreading to the whole population. But if this type of cancer can spontaneously originate, how do we avoid this happening a third time?

Unfortunately, I haven’t made it to Tasmania yet (did I mention I traveled to the other side of the world recently), but I did see a Taz devil in the zoo. Can we save Taz, so he may roam around and make weird tornado thingies?

Tazzie spotted in Taronga Zoo (Sydney)… Okay, they’re kind of cute in a giant rodent-looking type of way.

* Okay, technically speaking that means that it is still indirectly due to human influence.

Physics of Cancer (1)

If you are confused by the title, that’s okay. Usually, when we read something about cancer, it is about something biology-related, for example about specific mutations or the environmental conditions that increase cancer risk. A lot of research is happening with regards to the biology and biochemistry of cancer: which tumour suppressor genes are mutated in certain cancers, what are the effects cancer has on someone’s health, what drugs can we use to treat a cancer, … ? But, perhaps surprisingly, studying the physics of cancer also has its merit. Why, it’s a whole field in itself!

So I’d like to talk a little bit about this topic, the physics of cancer, and in this first part, I will focus on how physical forces can change the behaviour of cells (and how this might be involved with disease).

Cells not only sense their biological environment, they also feel their physical environment. They sense the stiffness of the cells and protein structures around them, they sense how other cells are pushing and pulling on them, and then they react to it. And these mechanisms could actually be quite important for the development and progression of cancer.

Recent research showed that the cells surrounding a tumour are under mechanical stress because of the growth of the tumour. As a tumour grows, it pushes on its environment. So the – initially healthy – cells in its direct surroundings, feel a pressure. In this specific study, they showed that this pressure caused the cells to start a mechanical response pathway leading to the upregulation of a protein β-catenin. This protein is involved in activating certain pathways involved in cell proliferation.

Which is exactly what its upregulation leads to in cancer. In the case of colorectal cancer (which, if you remember, I am particularly interested in), a mutation of Apc (adenomatous polyposis coli, in case you were wondering) also leads to an accumulation of β-catenin amongst other things. The APC protein has been linked to many functions, but the best known is its involvement in forming a complex that binds to β-catenin and tagging it for destruction. That way the proteins involved in protein recycling know that the β-catenin proteins can be cut up. But when APC is mutated, β-catenin gets tagged and starts piling up and doing some of its jobs a little bit too well, including inducing proliferation pathways.

So back to the study, if healthy cells are experiencing a constant pressure (due to a big bad tumour growing into their space, or – as they tested in the study – artificially caused pressure), they start acting more “cancer-like”. This suggests that mechanical activation of a tumorigenic pathway, in this case, the β-catenin pathway, is a potential method for transforming cells.

This is just one example of how physics and cancer are potentially related. As a side note, I myself am also interested in how cells respond the mechanical stresses, which prompted me to do an experiment where I placed weights on top of cells.


Feeling the pump.


This subject was the topic of my first FameLab performance, which ended in a little song (to the tune of “Friday I’m in Love” by the Cure). It’s sung from the perspective of a cell that is stuck next to a growing tumour:

Hello there, I am a cell.
Feeling healthy, fit and well.
Life is good, yes, life is swell.
But my neighbour’s got it worse.

Something about him does not belong.
The way he pushes is just wrong.
They say in him the force is strong,
they say he’s got the force.

He takes up so much space.
And is always getting up in my face.
It’s putting me in a stressful space.

You could say he’s left his mark.
It’s like swimming with a shark.
He’s pushing me towards the dark,
the dark side of the Force,
the dark side of the Force.

Oh, have a mentioned that I like Star Wars?