2016/05/31

San Francisco (1)

Last Monday was Memorial Day in the honor of all people who died while fighting for the US. Some cities have parades and it's a small 4th July. The direct effect on my schedule was, that I could take Monday off and drove with some friends to San Francisco which is about six hours by car away from Santa Barbara. SF is amazing and diverse city! Everybody warned me about the micro-climate in different parts of the city, but we only had sunshine and a bit of fog on Monday morning. I'll write more later, but here are some first pictures:



 (When we biked the Golden Gate Bridge!)


2016/05/26

Why we did it...

We finished up the project today - we've been working on it for many months. That means, that we will write up what we've learned about sea urchins, microvilli ("hair" on sea urchins) and confocal microscopy. And a paper? Well, we found out that there was no novelty in our research and other people have done what we did years ago with a much higher precision. However, we've learned a lot and we have at least one chapter for our theses. And maybe the understanding and learning we gained is even more important then a publishing a paper (though that would have been nice as well).

What we thought, we had discovered, is deconvolution of an image. When you take an image with a confocal microscope, you don't obtain a sharp image due to a couple of factors (e.g. resolution limited by the light's wavelength) and attenuation by the sample. Thus, we wrote a simulation of this process and understood where light is attenuated and how we can describe this mathematically as well. However, the de-blurring of images usually happens via deconvolution and non-trivial algorithms which leads to very impressive results:


Left: Image as it could have been taken with a microscope; middle: deconvolved image; right: "original" image / sample
 
Thus, there are experts out there who can do this so much better then us! But we gained personal understanding which will be useful for any further microscope imaging :-)

2016/05/23

iDrive - Matilija Falls

Have you ever confused the gas and break pedal in your car? So far, I didn't - but then the simplicity of an automatic car messed up my thinking about the number and locations of the pedals for a few milliseconds. I'm glad, nothing happened and I didn't get it wrong again.
There's no clutch in most American cars, so you don't need your left foot at all, which is kind of weird at first. Driving becomes very easy since you just drive or break. (Or "park" or "reverse" - those are the other two options for the gear stick, but you don't change the strength of the gear, just the mode)

To practice driving, we went hiking (yeah, lots of logic in here ;)) and it took us 1.5 h to get to the path (Matilija Falls) which was absolutely worth it - the water falls were just desiccated and a small creek was left:







But imagine the road in the mountains! Lots of steep slopes, small roads, no sight on the upcoming part of the road, serpentines and many steep curves, ... ideal conditions to practice! :)

2016/05/21

From uniform to Gaussian

On one of the homework sets for my statistical mechanics class, we have to juggle with probability distributions. Why's that interesting? First of all, particles don't behave random, but it would take too much computational power to track all particles in a system - imagine saving six values for 10^23 particles and then starting to calculate something. Thus, there are some fancy and useful probability distributions out there to describe a macroscopic system without calculating everything form macroscopic details. Furthermore, your computer needs random numbers all the time (in general, for encryption etc.), but also to run the simulations for those particle systems. Thus, we should care about probability distributions and I started writing the following blog entry:

It's somehow "simple" but there's still some magic about it: The Box-Muller algorithm transforms random numbers drawn from a uniform distribution in the interval [0, 1] to numbers which belong to a Gaussian distribution. What I mean: You draw two numbers from the interval [0, 1] and you might get any number from this interval with the same probability. You put those numbers into the Box-Muller algorithm and you get two numbers which resemble a Gaussian distribution, i.e. anything from minus infinity up to plus infinity, but it's most likely that you land near your mean value. How often you get close to the mean is defined by the standard deviation. For a detailed mathematical description you can look here.




Our homework asks for the derivation of that mapping, so how do you get to that idea? Well, first I thought of how I could get to any number between 0 and infinity from 0 to 1. The natural logarithm can do that for you; and the negative one gets you to any number from minus infinity to 0. You fix this sign change by a cosine and sine function and "that's it" :)

Lights in Isla Vista

We watched the Disney comedy "Zootopia" for only 4$ in the Isla Vista theater yesterday. It's comparable to the "Unikino" in Göttingen, but the movie choice is a bit more random and you sit in actual theater chairs and not in a lecture hall (so it's a bit more comfortable). Though "Zootopia" is a Disney production and "just" an animated movie, it's surprisingly good (actually, for all ages and levels of education).

It was about 10 o'clock when we got out of the theater and ran into a light exhibition in the student town. It was very nice and impressive:



2016/05/17

Visa renewal

If you receive a Fulbright Scholarship, you are automatically on a "Fulbright visa" (category "J" for exchange visitor, unlike most students who are in "F" which is the regular student visa). However, since your Fulbright grant only lasts a year, the visa expires at the same time and you have 30 days to leave the US (grace period). 

As a consequence, I have to go back to Germany in June, renew my visa and come back to study here until October! Basically, I fly all that way to get a new stamp. That's what the German Fulbright Commission told us.

As it turned out, there is a difference between your visa and your status document in the US. The visa is the printing in your passport, but your status as a student is depending on a specific form, the DS2019, which contains information about the university you stay at, which program you're in etc. The visa is only needed when you cross the border and not afterwards. In contrast to that, your DS2019 has to be valid and up-to-date all the time.

For that reason, the American Fulbright Commission updates your DS2019 very fast and easily, so you can stay in the US even though your visa is expired! Only, when you leave the country and have to re-enter, the passport control will expect you to have a new visa (it's similar to a cinema ticket, once you're in, it doesn't matter any more). I found out about the process way too late, so my flights are booked - but there's a conference I'll visit when I'm in Germany which is worth flying back for.

2016/05/15

Kayak your way to the Californian spirit

Finally we managed it to rent a three-person kayak from the Excursion Club and to get out on the ocean! We were only two people, but we filled the kayak and had a great time. The water is cold and the waves directly at the beach are strong, but once you're out there it's just beautiful and you can sun bath in your kayak. It's hard to turn the boat which is good since you do not want to get in touch with very large and scary see wheat or natural oil:





2016/05/13

Lunch and learn

Do you know what research in Education Science means? To be honest, I had no clue and I was happy to find out at "Lunch and learn" which is a series of talks where graduate students from all fields talk about their current research or present a paper they're familiar with. It's only happening about once a month, you get free pizza ;) and it's a great atmosphere since the presentation is in a rather small seminar room and not in a lecture hall.

By the way, that's the entrance to our new library where the lunch and learn seminar took place. It's got lots of study space and it looks pretty fancy.

As mentioned, last time a graduate student who's studying Education gave a presentation with the title "The solution to tragedy is community". First, he told us his personal story about one of his summers where about four of his close relatives / close friends died. He found comfort in the remaining part of his family / friends and got over it. This tragedy happened to an individual, which is something different to a tragedy that's striking an institution, like the UCSB.

In 2014, there has been a shooting in Isla Vista, which is the student town directly located next to the campus. Six people died and fourteen were injured. During Halloween 2013, the police arrested about 250 people due to riots in Isla Vista. These events and some others lead to a drop of 15% in the number of applications to UCSB.

The graduate student at "Lunch and learn" compared the portfolios of the UCSB before and after the shooting, talked to professors and students about these events and overall studied how the UCSB's identity was re-shaped to cope with the "dangerous place" stamp. His studies show so far that many posters, catalogs and interviews focus on community, i.e. that 75% of all students live either on campus or closer than 1 mile away, there are many student organizations and clubs, lots of recreational sports etc. I saw his point, but I think, that community is probably not that unique to Santa Barbara; yet, it seems like it has helped UCSB to go back to a "normal" university.

2016/05/10

Even more elephant seals

Yay, I fixed my laptop with a screw driver (that qualifies for the next post ;)) and now I can finally upload more pictures from our micro-roadtrip to Hearst Castle and the elephant seals:
So, CLICK HERE! :)

Hearst Castle was built between 1919 and 1947 by the architect Julia Morgan with and for William Randolph Hearst who had inherited a fortune made from gold-mining and who became a newspaper magnate. William Hearst collected lots of (large) art pieces in the Castle, shipped in animals from all over the world for his private zoo and also broad parts of a Roman temple to decorate one of his three swimming pools. It's a very impressive sight and it's worth seeing. Today it's only used for educational and touristic purposes (back in the day, it was some business/leisure convention center).

2016/05/09

Elephant seals

It takes only two hours by car to get to one of the tourist attractions in California, the elephant seals which give birth to their children at the Californian coast, roughly between Los Angeles and San Francisco. Hundreds of these animals basically lie on the beach and do nothing else then throwing sand at themselves from time to time (but not too often, that might become stressful ;-)). Indeed, they resemble elephants in some way since the males have a very large nose and in general, they are large (3-9 m!) and heavy (1000-5000 kg!):



In total, five people went, but only one of us was older than 25 years, so effectively we only had one driver: Renting  a car is about 30$ a day without any insurances, if you are 25 years or older. But below 25 you have to expect the doubled amount! Thus, only one person was driving all way, but since the highways are long and broad,  it's rather easy to drive. We basically turned right twice within 100 miles to get to the elephant seals.

2016/05/05

Don't get sick

Last year I got very sick and had to vomit several times in a few hours. So I managed it to get to one of the Health Centers around here (single physicians seem to be rare) and after two hours of waiting time a medical doctor (MD, unlike Dr. in Germany) saw me for maybe 15 minutes - and I was charged 130$ for the diagnosis: normal stomach indigestion. Treatment and medication: Isotonic drinks and salt sticks. 
The weirdest part is, that I have three insurances and none of them covered the incidence. One is my normal German insurance, another on is included in my scholarship and the university makes sure that every student has an insurance as well. The insurances cover a lot, but that's only in case you go to hospital or the costs of treatment are above a certain amount, e.g. 300$. But as a normal student, how often do I end up in hospital and how often do I catch a cold? Of course, the costs of a hospital visit outnumber my minor complaints, yet, I guess that this happens a lot more often.

If you come to the Health Center and you're not an emergency, you are sent to the Advice Nurse who can prescribe simple medications, take tests or lets you go to an actual physician. This makes the system apparently more efficient and reduces wait times.

However, I am not too upset since we have a Health Center on campus and a few days ago I was sick again and I got treated after 10 min waiting time (since it was the University Health Center I didn't have to pay... yeah, this system is a bit different). They also sell common medication for lower prices then pharmacies due to student discounts!

Recently, my colleague (who is also an international student) went to the dentist and apparently the dentist was not too happy with the teeth he saw. Instead of calculating the total costs of the treatment, the dentist asked my colleague when he would fly back home to go to the dentist - the costs of the flight (he comes from the Middle East) plus the dentist fees would save my colleague 1000$ compared to the US dentist costs only!

2016/05/02

We'll get there

Eight months ago: "Well, actually, I'd like to toss it."

Two weeks ago: "Okay, let's toss it."

One week ago: "Waaiiit... There's hope!!"

The state of our project was not the most promising one when I started. After three months we reached the state that we had re-invented the wheel successfully. Or at least it felt like. Two weeks ago we were close to finishing up and mentioning everything we did in our theses, but please let's don't write a paper and get as fast as possible to the next project.

BUT then we talked to some people who are experts in image processing and who have the same problem as we have it right now. Most people either ignored it or changed the project in a way that the did not have to care about it. The point is, that many people reconstruct shapes from fluorescent confocal images, but often not the absolute intensity value, i.e. the concentration of a certain protein at some depth in the sample. 

We try to do exactly this and we ran into some issues. But now we have a simulation which will help us to predict how certain signals in the sample are changed (e.g. attenuated) by the sample and it looks like we will be able to quantify the concentration of our protein :-)