Physicist at Large

Reading for November 18: Diffuse Optical Imaging and Mammography

2009-11-08T19:00:00.000-08:00

Overview:
Here's a short, animated introduction to the idea of imaging with diffuse light. In fact, the site that it's on is full of good stuff.

The more technical aspects are outlined in a still fairly simple form here.

Papers:
1) A review article on breast imaging with diffuse light.

2) An article on a clever trick for seeing through opaque objects with light.

Additional Material:
1) Here's an article specifically on using light to image breast tumors.

2) There's a company called Imaging Diagnostic Systems trying to market a laser-based mammography technique. I don't know how much success they've had in getting their devices into clinics, but here's an article on their work from two years ago. If you're thinking of doing a presentation and/or paper on optical mammography, this might be a good company to look up. Look for their papers, check out articles on them, maybe even contact them and tell them you're doing a class project and see if there's any technical information that they share with the public.

Optical Society of Southern California Meeting

2009-11-04T17:28:00.000-08:00

The next Optical Society of Southern California Meeting is next Wednesday, November 11. The topic is space exploration. Because of the holiday the school is closed and there's no journal club, but some of us will be going to the meeting. If you are interested in attending (it is FREE for students and includes dinner), please do the following:

1) Register here by Sunday.
2) Let me know so we can work out a car pool.

Optical Coherence Imaging: For November 4

2009-10-24T12:28:00.000-07:00

1) A few links explaining the idea behind optical coherence tomography: A basic overview with a diagram. Pretty technical, but it's hard to find good stuff on this. A lot of the review articles that I found either have equations or are subscription-only. The wikipedia article seems to be OK, but this isn't my research field, so it's hard for me to say for 100% certain.

Here's one really good review article that I found.

2) A paper by Dr. Barbara Hoeling (who will be speaking on Nov. 4) on Optical Coherence Microscopy.

The basic idea of optical coherence imaging is to have two pulses of light interfere with each other. One of them bounces off of a mirror set at a fixed distance, and the other bounces off of a tissue. The pulses of light only interfere constructively if they travel the same distance. Because of the way these pulses are set up (I'll explain the concept in class) they don't interfere constructively if the difference in distance is lambda. It has to be zero. So, by playing with the distance, you can get an image that only has light that traveled a particular distance, e.g. only light that traveled 1 mm into the tissue. This is great for 3D imaging of the retina, for instance, and is widely used for that purpose. There's also work on using it to image arteries and blockage in arteries, although that stuff hasn't made its way to the clinic yet.

Two classes for spring

2009-10-21T16:53:00.000-07:00

1) Dr. Abramzon is teaching Physics 410, Biophysics. This course is a combination of lecture, seminar, and journal club, focusing on biological applications of physics. This will cover all areas of physics, not just optics.

2) I will be teaching Physics 234, which will cover optics and relativity. We will focus on the mathematical theory of optics rather than the biological applications that we discuss in journal club. However, an understanding of the theory is needed to design the biomedical devices that we're talking about.

Beating the Diffraction Limit: Paper for October 28

2009-10-21T11:55:00.000-07:00

Next week we will discuss ways to beat the diffraction limit. The basic idea is that a lens can't focus light down to a spot smaller than the wavelength of light, which means that it can't see anything smaller than the wavelength of light either. However, people have found ways to turn on light-emitting molecules just a few at a time, and thereby see things smaller than the wavelength of light. This article summarizes a lot of different approaches to solving that problem.

A few things to note:
1) There are a lot of acronyms. Similar things will get different names. If you see STORM, PALM, iPALM, FPALM, PALMEYRA, they all involve molecules that switch on and off randomly. If you see STED, GSD, or RESOLFT, they all involve molecules that switch on and off controllably (i.e. a laser beam determines if they switch on or off).
2) All of these things require using special molecules. The lenses are still the same.

This is what my research students and I work on.

Neuroimaging: Maybe useful for people doing projects

2009-10-14T12:57:00.000-07:00

The latest issue of Nature has an article on imaging the nervous system with light. For those of you looking for a project topic, it might be good to skim this, find some topic that you find interesting, and then check out the references.

Fiber-Optic Biosensors

2009-10-13T14:01:00.001-07:00

Next week Dr. Salik will talk about fiber-optic biosensors. Read this article to prepare. The article talks about several sensors, but Dr. Salik will focus on fibers. In the first few pages, just skim for basic concepts regarding sensitivity and applications, but don't worry about the detailed physics of plasmon resonance or photonic crystals. Focus on fibers and waveguides, since waveguides are very closely related to fibers. Also try to read up on Mach-Zender interferometer sensors if you get the chance, because they are related to Dr. Salik's work.

As good as a textbook

2009-10-09T17:02:00.000-07:00

The optics catalog published by Melles Griot has an excellent series of teaching materials, explaining lenses and imaging and beam optics. Take a look here.

Seeing through tissue with light to do mammograms

2009-10-07T18:17:00.001-07:00

Via Tony Durkin's blog, I learn of some new work on doing mammograms with light instead of x-rays. This concept has been around for a while (a cynic might say that it's the method of the future--always has been and always will be!) but it's always nice to see progress.

Good explanation of Confocal Microscopy

2009-10-07T17:52:00.000-07:00

Courtesy of your classmate Oscar Ledezma, here's an explanation by Eric Weeks of Emory University. I don't know Eric Weeks, but I know people who know him. Science is a pretty small world.

Physics Department Fall Social

2009-10-07T11:47:00.000-07:00

If you're a physics major, physics minor, or just plain interested in talking to people in the physics department, come to our fall welcome social next Tuesday from 11am to 1pm in 4-1-314.

Confocal Microscopy: For October 14

2009-10-06T11:15:00.000-07:00

After we finish reading the paper on two-photon microscopy, we'll start reading about confocal microscopy. Here are some good papers on it:

1) A short editorial on confocal endoscopes.

Honestly, the most important thing from this article is the figure. It gets the concept across better than most of the stuff I've seen.

2) Two articles (written several years apart) by the same research group on a confocal endoscope. We'll compare the articles to see how they made progress. Start with the more recent one, as it's easier to understand.

Items 1 and 2 are what you should focus on. If you read those 3 papers, that's enough.

For additional info:

3) Nikon Microscopy U. has a few articles. They're OK, but not spectacular.

4) American Journal of Physics article on how to build a confocal microscope. The American Journal of Physics is a great place to look if you want to get info on how to do something for a senior project.

If you want to write a paper detailing what you'd have to do to build a confocal microscope on campus for a senior project, this is a good paper to start from. I'd love to read a paper detailing a possible senior project.

5) A medical review on confocal microscopes in the clinic. Don't read this one in detail, just skim it to get an idea of how important this stuff could potentially be.

If you want to write a paper on medical applications, this article (and the first one mentioned in this post) are both excellent places to start.

Reading for October 7

2009-09-30T18:12:00.000-07:00

1) Read the 2-photon paper.

2) Start reading the Nikon Microscopy U and American Journal of Physics articles on confocal microscopy.

Extra credit for 132 students

2009-09-27T14:34:00.001-07:00

In order to encourage freshmen and sophomores to learn what sort of things physicists work on, anybody in my physics 132 lecture who attends the Wednesday journal club discussion will get 10% extra credit on homework if after the discussion they send me a short paragraph describing what we talked about and what they learned.

First Reading: Two-photon microscopy

2009-09-15T14:54:00.000-07:00

In the first meeting (Wednesday, September 30), after going over a few admininstrative matters we will talk about the following:

1) How to find biomedical optics papers online? For those of you planning to give a presentation and/or write a research article, this will be useful.

2) How to read a paper? As an example, we will examine this paper on 2 photon microscopy. Read as much of it as you can, and don't be afraid to skip confusing sections. We'll talk about the basic idea and the key things to look for in a paper.

We'll probably continue the discussion of this paper in the following week.

Things to keep in mind as you read the paper:
1) It's often productive to start with the diagrams.
2) Don't be afraid to skip stuff that you don't understand.
3) This is a review article, which means that it provides an overview of the field rather than discussing a specific experiment. Review articles tend to start out general and become more specialized. Don't be afraid if you find the later parts harder. (However, if you're a biologist, the last couple pages might be more interesting, while physicists and engineers will find the first few pages interesting.)
4) Another good introduction to this topic is at MicroscopyU. MicroscopyU is full of good stuff, and you should always consult it when you are reading an article and run across things that you don't understand.
5) In the first few pages of the paper, especially in box 2, you'll run across a "point spread function." MicroscopyU has an OK explanation of it. Here's another OK explanation. The Wikipedia article is horrible. We'll talk about this concept in class.

The basic idea of fluorescence microscopy is simple: You illuminate an object, and wherever it's illuminated it produces light. You'd like it if most of the light you detect comes from wherever the lens focuses the light, but sometimes (especially in tissue that scatters or distorts light) this isn't possible. Basically, light has to travel through other tissue to reach the focal spot, so the other stuff produces light too.

Two-photon microscopy gets around this by using a trick: For special dyes, the amount of light they produce is not proportional to the amount of light they get. Rather, it's proportional to the square of the amount of light it gets. (What this has to do with two photons will be explained in class.) So there's only a strong signal from the spot where the light is focused, and you detect that.

Anyway, we'll probably spend a week and a half reading this, because it introduces lots of important ideas related to imaging. Then we'll spend a week or two talking about confocal endoscopes. After that, we'll spend a week or two talking about new tricks (including tricks that my students and I are working on) for making microscopes with resolution that scientists used to think was impossible. We'll also have presentations by a few other professors, and finish the quarter with a few weeks of student presentations.

Syllabus

2009-09-15T10:32:00.000-07:00

The syllabus and schedule are posted here. The journal club will meet Wednesdays from 4pm to 5pm in the building 4 second floor conference room (4-2-314). The policy for credit is:

If you wish to receive 1 unit of credit, you must do the following:
1) Register for Physics 400
2) Attend at least 7 meetings (with exceptions for unusual circumstances)
3) Participate in the discussions
4) Lead a discussion of a paper near the end of the quarter.

Grading will be: 50% participation, 50% leading a discussion.

More specifics on grading:

Participation: You’ll get half credit if you participate, and full credit if you come prepared with some questions from the reading.

Presentation: 1/3 for preparing a good summary (paragraph) and reading guide (a few bullet points) for people to read with the paper (due the week before your discussion), 2/3 for showing what you understand in the paper (and articulating what parts you aren’t sure about and why).

A: 90% or better
B: 80%-89%
C: 70%-79%
D: 60%-69%
F: Below 60%

If you wish to receive 2 units of credit, you must do everything listed above for 1 unit of credit, and also write a paper (8-10 pages). In that case, the paper will be 50% of the grade, and participation and discussion will each be 25%. The paper can be related to the topic that you present on. (In fact, I strongly encourage you to do that.)

New class!

2009-09-14T15:42:00.001-07:00

This fall I am using the blog for a biomedical optics journal club. Since Physics 344 (Applied Optics) was canceled, this weekly journal club will meet to read and discuss papers on medical applications of optics.

More information coming soon.

Clickers

2009-05-12T18:07:00.001-07:00

This is a thread for discussing clicker questions. Note that the clicker questions themselves are found here.

Homework discussion

2009-05-12T17:48:00.000-07:00

This is a thread for discussing homework problems.

Honors Section Online Office Hours

2009-03-11T10:49:00.001-07:00

Post your final exam questions in here at any time. I will be in this thread for real-time chats on Monday, March 16 from 11am to noon Pacific time (2pm-3pm East Coast).

Final Exam Online Office Hours

2009-03-11T09:41:00.000-07:00

I will be out of town March 16-18. However, on Monday, March 16, from 10am-11am Pacific time (1pm-2pm Eastern) I will take questions in the comments section of this thread.

Homework 3 discussion

2009-01-19T14:43:00.001-08:00

The comments section of this post can be used to discuss homework 3 with other students.

Tutorial 3 discussion

2009-01-19T14:42:00.001-08:00

The comments section of this post can be used to discuss tutorial 3.

Homework 2 discussion.

2009-01-19T14:41:00.000-08:00

Sadly, I can't always answer emails the moment they arrive, and I can't always be there to conduct extended email chats about homework. If you'd like to discuss homework with other students, the comments thread of this post is available for that purpose.

Tutorial 2 Discussion

2009-01-19T14:37:00.000-08:00

Unfortunately, I cannot always answer email the second it arrives. So, if you'd like to discuss tutorials in here, post your questions as comments.

Old Files

2008-12-01T16:35:00.000-08:00

I've been reorganizing my directories. A lot of files for this class got moved to this directory. Links in old blog posts won't work (don't worry, the syllabus and latest homework are fine) but anything from old assignments can be found in the new directory here.

Last homework: Ising Model

2008-12-01T16:25:00.000-08:00

Everything you need for the last homework (due Dec. 10 at 6pm) is in this folder.

Problem 2 on last homework: How to test it

2008-11-26T17:52:00.001-08:00

Here's the easiest way to check your code for problem 2:

Create an array of ones:

spins = ones(20,20);

The energy of that system should be negative because all of the spins are aligned (minimum energy configuration). Then, tell the function to calculate the energy change if a spin somewhere in the middle is flipped. The energy change should be positive, because flipping a spin increases the energy of the system (it's now opposed to its neighbors, when it wants to be parallel with its neighbors). After that, check the energy change for a spin at one of the corners. The change should be positive but half as large as before, because that spin has only 2 neighbors rather than 4. Finally, check the energy change for flipping a spin at one of the edges (e.g. x = 1, y = 10). That energy change should be half-way between the energy changes calculated at the center and at the edge, because there are 3 neighbors.

Syntax for making graphs

2008-11-19T17:22:00.000-08:00

Problem 3 on the latest homework requires you to make a plot. I recommend using the following commands: (entered from the command line in Matlab)

figure; surface(array name, 'edgecolor','none'); axis equal; colormap jet;

'edgecolor', 'none': This tells it whether to put in grid lines. Try it without the 'edgecolor', 'none' and you'll see that it's hard to get a good picture--the grid lines obscure a lot of info.

axis equal: This tells it to make the horizontal and vertical directions on the same scale. The picture displays better that way.

colormap jet: This tells it the color scheme. Look at the help files on colormap and try a different color scheme if you want to play around a bit.

Grading Update

2008-11-11T15:23:00.000-08:00

I originally promised you 6 homeworks, i.e. 6 chances to improve your grade. However, homeworks 2 and 3 wound up taking more time than anticipated. In an effort to make the impact of each homework proportional to the time spent on it, homeworks 2 and 3 will be counted as out of 30 points in my grade spreadsheet. I'll still grade them on a 20 point scale, but I'll multiply those 2 grades by 1.5.

Example files: Random Walks

2008-11-08T21:16:00.000-08:00

Here are 2 example files:

WalkEm.m: This file simulates a simple random walk.
WalkEmForward.m: This file simulates a random walk that doesn't double back on itself.

Homework 4: Random Walks--UPDATED

2008-11-08T14:27:00.000-08:00

Homework 4 is available, and it's due Nov. 26 (day before Thanksgiving).

I made a small change to problem 3. The photon now starts off at y = 2, not y = 1. This gives you slightly better data and it's physically realistic: Assume that we used a lens to focus the light to a point somewhere below the surface of the tissue rather than at the actual surface.

Homework 3 tip

2008-11-05T16:51:00.000-08:00

For the first part, where you're calculating sizes, do the calculations to 0.01 microns (10 nm). This is a tolerance that is achievable in the lab.

Waveguide Example File

2008-11-05T13:52:00.000-08:00

Here's the example file I'll be using in today's class.

Extra Credit

2008-11-03T17:43:00.000-08:00

Give me a useful critique of this research plan by Sunday, November 9, and I will give you up to 5 points of extra credit on Homework 3 (tentative due date: November 12). If the proposal is funded, I owe you dinner at the Cold Springs Tavern in Santa Barbara.

Homework 3 was revised

2008-10-29T18:05:00.000-07:00

The third homework has been revised to correct a typo. It is here.

Note that on problem 2 part a, you won't get perfect agreement between the numerical result and the exact solution, but they should be pretty close. Changing the boundary condition at r = 20 will help a little bit.

Example file: Diffusion Equation

2008-10-22T15:24:00.001-07:00

Here's a file that we'll be using to solve the diffusion equation.

Homework 3

2008-10-21T20:49:00.000-07:00

Here it is. It's due on Monday, November 3.

EDIT: That due date will probably be extended.

This is perhaps a better laser article

2008-10-21T19:24:00.000-07:00

The article linked to here provides much of the motivation for problem 1 on the homework.

Believe it or not, this is a good article on lasers

2008-10-20T15:36:00.000-07:00

There is a website called "The Britney Spears Guide to Semiconductor Physics." It seems to have been written by a bored graduate student who (for some strange reason) likes Britney Spears. Leaving aside a few of the more questionable pictures (none of them on the page I'm linking to) the physics content there is actually pretty good. So read this article for more on edge-emitting semiconductor lasers.

Homework due date changed

2008-10-15T17:56:00.000-07:00

Homework 2 will be due on Wednesday, October 22.

On Monday we will go over remaining homework issues, and then I will lecture on boundary value problems.

Useful Command

2008-10-15T16:50:00.000-07:00

Subplot is a very useful command for putting more than one graph on the screen at a time.

The help file is here.

Examples: Differential equations

2008-10-13T21:24:00.000-07:00

Here are 3 useful things:

1) I revised the function that calls the Runge-Kutta Solver. I changed it so that the times and array sizes are handled properly. Before, the loop started from t1, which wasn't very sensible because the arrays times and yvalues had already been initiatlized with t1 and y(t1). Now it starts from t1+h and continues to t2-h. Why does it end at t2-h rather than t2? Because it uses y(t2-h) to figure out y(t2).

2) I also give you an example of how to modify the function f to handle more than one variable. Let's say we want to solve the equation d^2 y/dt^2 = -y. We can define two new variables y1 and y2, and our equations are dy1/dt = y2 and dy2/dt = -y1. The array "yold" contains the old values of y1 and y2, stored as "yold(1)" and "yold(2)". The array "timederivative" contains the derivative of y1 in its first element, and that derivative is "yold(2)". The second element of "timederivative" is the derivative of y2, which is "-yold(1)".

The complete text is here:

function timederivative = f(yold,t, params)
%Here's where the physics is

timederivative(1) = yold(2);
timederivative(2) = -yold(1);

end

3) If you want to display the output of commands in scientific notation, run the following Matlab command:

format short e

To test it, type "pi" at the command line and hit enter. You should see " 3.1416e+000."

Homework 2: Revised

2008-10-13T21:15:00.000-07:00

Here is the revised homework 2.

Euler files

2008-10-06T17:26:00.000-07:00

The files "EulerSolver" and "RunEulerSolver" are here. You should modify them to run the 4th order Runge-Kutta algorithm.

Second homework

2008-10-06T15:03:00.000-07:00

Here it is. Due October 20.

Don't ever, EVER trust Wikipedia for math

2008-10-02T16:45:00.000-07:00

Wikipedia is a great place to get started in learning about lots of things, but don't EVER trust an equation on Wikipedia. I've found errors in them, and typically very subtle errors that a non-expert would miss.

Trust Wolfram MathWorld. Don't trust Wikipedia for math.

Matlab Logic

2008-10-01T17:36:00.000-07:00

[EDITED IN RESPONSE TO COMMENT. I'M MIXING UP THINGS FROM DIFFERENT PROGRAMMING LANGUAGES.]

A few points about logical statements in Matlab:

A statement like "If x equals 5, then do this" is NOT implemented as "if(x=5)". It returns an error. Matlab takes the statement "x=5" and evaluates it. The output of that statement is a non-zero positive number, and to Matlab that non-zero positive number is equivalent to TRUE. Even worse, in evaluating "x=5" Matlab automatically sets the value of x equal to 5. So your statement is always true. That's no good.

Instead, you should implement this as "if(x==5)". To Matlab, a double equals sign is a logical operator, testing whether or not x is equal to 5, whereas a single equals sign is an assignment statement.

Also, suppose you wanted to do something when x equals zero. If you wrote "if(x=0)" Matlab would RETURN AN ERROR. take that statement x=0, assign the value 0 to x, and then say "Hmm, the value of this statement is zero, which means FALSE" and in the process you have (1) guaranteed that the value of x will be zero (even if it wasn't supposed to be zero) and (2) you have guaranteed that it will NOT do what it was supposed to do when x is zero. So that statement SHOULD be written as "if(x==0)".

What about statements like "IF x is NOT equal to 5"? For those, write "if(x~=5)".

Multiple conditions: "&&" is "AND" and "||" is "OR". So write "if(x==5&&y==6) for "if x equals 5 and y equals 6" and "if(x==5||x==5)" for "if x equals 5 OR x equals 4."

Interesting Physics

2008-09-30T10:29:00.000-07:00

Perhaps you've heard of "negative index materials" that might be used to make invisibility cloaks that bend light around an object. Well, they've just come up with a version of that for water waves. Summary here. Original article here.

A scale model of this would actually be pretty easy to build and test with a ripple tank. We have one in the department.

Homework 1: Intro to Matlab and data handling

2008-09-27T21:36:00.001-07:00

The first homework assignment is here (.pdf). The matlab file for problem 1 (which you will have to modify and send back to me) is here. The data file for problem 2 (which you will have to analyze) is here.

I would appreciate if you post your questions in the comments thread. Even if you have to email me files, at least put the substance of the question in the comments thread (and reference the email you sent me) so that other students can see the discussion. However, if you prefer to keep your questions private, you are still free to email me.

Least Squares Resources

2008-09-27T21:33:00.001-07:00

In the second lecture (October 1, 2008) we will talk about least squares fitting as an example of handling data in arrays. Wolfram MathWorld is generally a good place to go for math info, and so I'll put a link to their least squares article here.

Files from first lecture: 9/29/2008--UPDATED

2008-09-27T21:31:00.001-07:00

All of the files that I'll use in my first lecture can be found here.

After class I will try to upload a history file as well, so you can go through all of the commands that I typed in Matlab.

UPDATE: History file has been added.

Change to course

2008-09-17T11:16:00.000-07:00

Prof. Siegel's schedule prevents him from being able to teach computational physics this fall. I'll be the only instructor for the course.

L33T Einstein PWNS N00bz!

2008-06-10T11:14:00.000-07:00

There is now a computer game that lets you fly through an obstacle course near the speed of light and see how things look different because of relativistic effects. You can even look at clocks and see them run at different speeds. It's done by a professional game programmer in collaboration with a physicist, so the graphics are good.

I don't think there any aliens to shoot at, unfortunately.

Final Exam Questions

2008-06-02T16:04:00.000-07:00

UPDATE: Somebody just pointed me to an error in the solution that I posted for problem 34 from Chapter 18 (homework 7). I can't prepare a new solution in neat form tonight, but here's what I wrote to that student as advice on the problem: 1) Mean free path won't be on the final. 2) For thermal energy: You are given P and V. Thermal energy is n*Cv*T. It's a diatomic gas, so Cv = 2.5 R. Eth = 2.5*n*R*T, and P*V = n*R*T. 3) After expanding, you have a different value of P but the same volume V.

Hope this helps.

NOTE: Online office hours are over, but I'll continue to check for questions periodically. I'll be at conference presentations most of the time (i.e. I won't have my laptop out) but I will check periodically.

This is the thread to post any and all questions related to the Physics 132 final exam. I'll answer them in comments as they come in. I will also camp out here for 2 hours on the afternoon of Tuesday, June 10, from 11am to 1pm West Coast time (2pm to 4pm East Coast time, where I'll be). In that time you can ask me questions in more of a discussion mode.

Reminders regarding final exam:
1) It is comprehensive, covering all of the chapters that we covered in class.
2) You are allowed 2 pages of notes.
3) You are allowed to use calculators.
4) The exam is 1:40 pm to 3:40 pm in room 8-4 on Wednesday, June 11.

So to ask a question just post it in the comments.

EDIT: Somebody asked me for a diagram of the problem involving a tunnel through the earth. There won't be a problem in that exact same category on the exam, but solving it illustrates some general concepts that are important, so I'm putting a picture of the solution here.

Paper just published!

2008-06-01T10:24:00.001-07:00

If you want to know what I do, check out my latest paper.

I only read it for the articles

2008-05-20T11:19:00.001-07:00

Being a biophysicist interested in blood vessel development, this is actually relevant to my work.

I must be doing something wrong

2008-05-16T17:56:00.001-07:00

A friend got me interested in a side project. It appears that the only way to solve this problem is to delve into the properties of Bessel Functions. Clearly, I have erred somewhere along the line. Bessel functions are nasty things, not nearly as elegant as exponentials and trigonometric functions. Properties of exponential functions and trig functions can be figured out by hand, and in doing so you get some insight. With Bessel functions, you have to consult tables of identities.

Surely I did something very sinful to incur this punishment....

"Thank you, Mario, but our protein is in another configuation."

2008-05-09T12:42:00.001-07:00

University of Washington researchers have developed a computer game that folds proteins. Yes, really.

Protein folding is a tough problem. You've got this long chain of amino acids, and somehow they come together to take on a particular shape. And they do this every time a cell produces the protein. You'd think that a long chain of molecules would try all sorts of random configurations, but somehow they always form the same pattern. It's hard to understand why, because all of the molecules are interacting with each other in very complicated ways, so you have to solve an equation that could involve hundreds or thousands of forces. Some computers are having luck folding a few proteins, but it's slow.

The researchers theorized that people are very good at playing games and picking out the most efficient way to do something. Good gamers intuitively know which traps to avoid. So they designed a game where you try to fold up a molecule into the most stable shape (the one where the forces all cancel and the energy is at a minimum). They tested it on some proteins with known structures, and they got good results. Now they're trying proteins with unknown structures.

So to all my students, I say "Go and play more video games! Do something useful for science!"

The Physics of Melting

2008-05-08T10:18:00.001-07:00

Here's a nice synopsis of an article on the physics of melting. Since I mentioned melting yesterday in class, I figured some of you might find it interesting.

Testing my cell phone blogging.

2008-05-07T23:07:00.001-07:00

Testing my cell phone blogging.

EDIT: The way it's set up right now, I can send short posts from my phone as text messages, but I can't read it from my phone. I have a cheap plan, so I don't have web browsing on my phone. But I like being able to text short posts in case, um, well, I dunno. I'll think of some reason to text short posts from my phone.

Oh, I can send images. Yeah, I'll send those, some time.

It's embarrassing to be far less wired than my students.

How to invite an alien invasion

2008-05-07T22:36:00.000-07:00

Ever worry that it's a bad idea to beam radio signals into space in hopes of making contact with aliens? Ever worry that the creatures getting the signals might not be benevolent? Well, fear no more: If aliens detect our electromagnetic transmissions, the signal they're most likely to detect is the radar that we've been using to search for planet-destroying asteroids.

What do I plan to do with this site?

2008-05-07T18:56:00.001-07:00

Here are my plans for the site:
1) Blog science news items, so that students (and anybody else interested in physics) can see what's going on outside class, and get my take on it.

2) Continue the discussion outside class. I won't implement this feature in my current quarter's classes (better to do it from day 1), but in the fall I see myself blogging thoughts on things that I didn't explain fully in lecture, or responding to points raised in class that merit further discussion.

3) In smaller classes, give students a place to guest blog their thoughts, to better continue the discussion. Whether it's a bleg for homework help or a draft of a paper, guest blogging privileges will be a way to continue the discussion outside class.

4) In advanced classes on specialized topics, bring in guest bloggers who are experts in the field (especially people teaching at schools other than Cal Poly) to offer their own thoughts on whatever we're discussing.

For now, though, I'll mostly just do step 1, to build the readership. As promised, of course, I will hold office hours in a comment thread when I have to be away from campus for a conference.

+1

2008-05-07T18:31:00.000-07:00

Here's a nice post on the way that physicists all too often pigeonhole themselves and others, especially when we interact with biologists.

The reality is that if you're trying to apply fundamental principles to get a deeper understanding of a process in nature, you're doing good science and having fun. However, Philip is correct to point out that just because you're working on a biological system that doesn't necessarily mean you're getting to the heart of the biological issues. Anyway, have fun in science, and don't worry about the labels, just don't fool yourself into thinking that once you move past labels you're doing somebody else's work "the right way."

Those chemists can do anything

2008-05-07T18:17:00.000-07:00

Well, since my students are apparently reading this blog (or at least one of them is), I might as well start blogging items that I enjoy.

Today's item is "Point and Click Chemistry." I just learned of this concept today, but I always enjoy learning what chemists can do. Most of materials science and biophysics is utterly reliant on the ability of chemists to continue to produce neat new tools and substances for us to work on.

What's particularly nice about this is that (1) they can fluorescently label things (i.e. attach a section to the molecule that will absorb light of one color and re-emit it with a different color, so people with microscopes can tell where different molecules are) and (2) they are targeting sugars. In the body, sugars do a lot more than just provide energy. They are a major component of the substance between cells, and the composition of that substance tells cells where to go and what to do. This is an important aspect of my research, where I try to develop mathematical and computational models of what cells are up to as they migrate to form blood vessels.

First post, and last post for a while

2008-05-05T16:22:00.000-07:00

Welcome to my blog. For now, it's just a place that I'll occasionally use for course-related purposes. Most of my professional information is posted at physicistatlarge.googlepages.com. My personal blogging is done under an alias at another site. Most of my online instructional materials are delivered via Blackboard or Mastering Physics, but occasionally it is convenient to have this blog available. Part of the reason is that those course management systems have expiration dates for courses.

The first anticipated use of this blog will be in June during finals week, when I'll be away from campus and hold "office hours" in a comment thread. For those who are wondering why I don't just use Blackboard's discussion boards for that purpose, that course's materials are mostly delivered via Mastering Physics, and Mastering Physics doesn't (to the best of my knowledge) have chat capabilities. Setting it up in Blackboard would be a bit more cumbersome than I like...ok, maybe not, but I like the blog format and plan to use it down the road for future courses.

Anyway, see you soon in a comment thread. Perhaps I'll also post the occasional physics news story here.