October 2007 Archives

Interview

| 7 Comments

Since I am in the second year of my post-doc, I’ve begun to apply for tenure-track faculty positions. Roughly, the jobs that I am interested in are at research universities and are looking for a statistical geneticist or computational biologist. I’d also like the opportunity to teach undergrads and have grad students. There are a lot of good ones out there this year, especially on the east coast.

Well, today I got my first offer to do an on-site interview for a position. I’m going to keep most of the details private, but I am excited at the opportunity.

While working to solve problems with PT and this blog, I made some improvements to MT-Dispatch since version 1.4. I’ve decided to produce version 1.5 with the new changes, while I wait for the cluster to finish my simulations. (Ancestral selection graphs are cool!)

Download version 1.5.View documentation.

Changes in this Release

  • If PERL_UNICODE is set, MT-Dispatch will decode all CGI parameters under the assumption that they are UTF-8 strings. The parameter ‘file’ is skipped because it is used for uploading binary data.
  • After a worker is finished serving the CGI page, it will run any waiting periodic tasks. This allows one to use the background publishing queue without running the periodic daemon.
  • The default maximum requests is now 100.
  • If PERL_FCGI_LOG is not set, stderr will be used instead. (Suggested by Hirotaka Ogawa)

A Discovery Institute Mascot?

| 2 Comments

lolcats and funny pictures

Ranking College Football Teams

Three years ago I fooled around with doing my own computer rankings of college football teams. I’ve decided to try it again, this time using some ideas from elementary kinetics.

The quality of each team is determined by its position on a line. Teams with better wins and better losses have a higher position than teams with worse wins and worse losses. All teams start out at position 0, and then move around the line based on the games they played, their scores of these games, and the quality/position of their opponents. This movement is modeled as follows:

  • is the number of points that team i scored against team j.
  • is the set of opponents that team i has played.
  • is the quality and position of team i.
  • is the velocity of team i
  • is the drag coefficient, which dampens the oscillations in the system.
  • is the net force being exerted on team i by its opponents. It is calculated by the following equation:

If is a small unit of time, then the velocity and position of team i are updated as follows.

From the starting position of all teams at 0, the system is allowed to run according to the above equations until it stabilizes within a certain tolerance.

Running the algorithm on the games played so far with and a stabilizes at iteration 2056. The full results can be found in this file. Below is a table of the top best fifty teams. Their positions have been scaled based such that the best team has a position of 100, and the worst 1-A team has a position of 0.

Rank Team Record Position Rank Team Record Position
1 LSU 7-1-0 100 26 Oklahoma St 5-3-0 66.76833079
2 Ohio State 8-0-0 91.91770206 27 Georgia 5-2-0 66.26334342
3 Oregon 6-1-0 90.37649059 28 Texas Tech 6-2-0 64.5558861
4 Florida 5-2-0 89.98650034 29 Clemson 5-2-0 64.43963901
5 Kansas 7-0-0 88.56028599 30 Michigan 6-2-0 63.79715507
6 Oklahoma 7-1-0 88.16654584 31 UCLA 5-2-0 63.37091573
7 South Florida 6-1-0 84.70038249 32 Illinois 5-3-0 62.57093573
8 Arizona St 7-0-0 84.64538387 33 Colorado 4-4-0 62.11719707
9 Auburn 5-3-0 83.03292418 34 Vanderbilt 4-3-0 61.96470088
10 West Virginia 6-1-0 83.01792455 35 Washington 2-5-0 61.9359516
11 Kentucky 6-2-0 79.32176696 36 Boise St 6-1-0 60.91472713
12 Missouri 6-1-0 78.85427864 37 Georgia Tech 5-3-0 60.79973001
13 Kansas St 4-3-0 77.79680508 38 Brigham Young 5-2-0 59.90100247
14 Arkansas 4-3-0 74.74063148 39 Michigan St 5-3-0 58.60603485
15 Rutgers 5-2-0 72.42568936 40 Virginia 7-1-0 58.51728707
16 Penn State 6-2-0 71.18947026 41 Florida St 4-3-0 58.44603885
17 Cincinnati 6-2-0 70.9919752 42 Texas A&M 6-2-0 57.41231469
18 Southern Cal 6-1-0 70.60698483 43 Louisville 4-4-0 57.29856754
19 Alabama 6-2-0 70.55448614 44 Purdue 6-2-0 56.90607735
20 Texas 6-2-0 70.34324142 45 Maryland 4-3-0 56.90232744
21 Virginia Tech 6-1-0 69.73700657 46 Tennessee 4-3-0 56.70733232
22 South Carolina 6-2-0 68.99577511 47 Miami FL 5-3-0 56.48983775
23 California 5-2-0 68.80077998 48 Oregon St 4-3-0 56.1398465
24 Connecticut 6-1-0 67.81830454 49 Wake Forest 5-2-0 55.73110672
25 Boston College 7-0-0 67.52581185 50 Troy 5-2-0 54.97737557

Now how do we know these rankings are any good? There is no “right” way to rank teams. However, there are two things that I think every computer rating system should be able to roughly do: 1) sort out the divisions and 2) sort out teams based on records. This ranking does both of these, so I’d call it a success. I still need to work on balancing blow-outs and shut-outs. I think I’m currently giving the offenses too much control over the rankings.

lolcat

cat pictures - Thumbcat  is evolving

Reposted from The Panda’s Thumb

As reported by NCSE, Sen. David Vitter (R-LA) has withdrawn his $100,000 earmark for anti-science education: Vitter earmark withdrawn. During yesterday’s session Vitter asked that the money in the earmark be directed to other projects, insisting that he never intended—yeah, right—for the money to go towards teaching creationism.

Joshua Rosenau of “TfK” has the full exchange on his blog.

Mr. VITTER Madam President, I rise today to discuss a project I sponsored in the fiscal year 2008 Labor, Health, Human Services and Education appropriations bill. The project, which would develop a plan to promote better science-based education in Ouachita Parish by the Louisiana Family Forum, has raised concerns among some that its intention was to mandate and push creationism within the public schools. That is clearly not and never was the intent of the project, nor would it have been its effect. However, to avoid more hysterics, I would like to move the $100,000 recommended for this project by the subcommittee when the bill goes to conference committee to another Louisiana priority project funded in this bill.

Mr. HARKIN Madam President, I appreciate the sentiments by the Senator from Louisiana and accept this proposal to move the funding for this project to other priority projects for the State of Louisiana in the bill when it goes to conference committee.

Mr. SPECTER Madam President, I concur with my colleague and will agree to move these funds in conference committee.

Congratulations to everyone who contacted their senators about this earmark. The students of Louisiana owe you a debt of gratitude. You not only got the earmark withdrawn, you also got the sponsor to do it.

Kudos all around.

Reposted from The Panda’s Thumb

Last January Prof. Steve Steve, Bora, and I met Congressman Brad Miller (D-NC) at the NC Science Blogging Conference. We took the opportunity to explain to him NCSE’s Project Steve and our Project Steve Steve.

Rep. Miller is the Chairman of the House Science and Technology subcommittee on Oversight and Investigation and promised us that he’d ask science skeptics that testify before him “How many are named Steve?”.

Today, he did it:

Do you all know any climate skeptics … named Steve?

During a hearing about the impact of global warming on the arctic he unexpectedly asked the panel if they knew any Steves that agreed with them. The global warming skeptics couldn’t name any off the top of their head, while the global warming researchers named two.

Miller also gave the Panda’s Thumb a plug for good measure.

The discussion about Steves begins a bit before one hour, fourteen minutes (1:14) in the RealPlayer clip of the hearings. It continues for a few minutes.

I’d love to see the clip on youtube if anyone can extract it. If not, a transcript would be nice. (A lot of people don’t like dealing with RealPlayer.)

I guess that Miller had a good time bringing up Project Steve because he directed his staff to email us about it.

Hopefully, more politicians will follow his example when dealing with “expert” testimony.

Since PZ has quite a storm going on in his blog about human “races” and intelligence, I think I’ll repost this great two-year-old post that just show how homogenous humans really are: Chimps are Laughing at You.

Given Nature’s publication of the chimp genome, I figured that I’d share one of my all time favorite figures from a scientific paper. It is from Gagneux et al. (1999) Mitochondrial sequences show diverse evolutionary histories of African hominoids. PNAS 96 (9): 5077.

Fig. 1. Unrooted phylogram of the neighbor-joining tree of 1,158 different CRI sequences before (A) and after (B) after topiary pruning to level PL = 8 to remove homoplasies. Bootstrap values >= 50% for the primary internodes are shown. Position of the midpoint root is indicated by arrow. Different colors indicate species (humans, bonobos, and gorillas) and subspecies (chimpanzees). Symbols indicate individuals belonging to the same social group.…

What is very obvious from this paper figure is how genetically homogenous we are compared to our closest relatives. I’ve heard it stated that a single chimpanzee tribe has as much genetic diversity as is found in the entire human species. So next time you feel fit to make fun of us southerners for kissin’ cousins remember that chimps are laughing at you.

State Fair

Tiffany and I went to the NC State Fair last night. She has the details.

I’ve also finally updated the layout and style based on some of my PT work.

When I started blogging three and a half years ago, I quickly realized that I needed some way to embed equations into my blog posts. No solution existed at that time; although, there were several implementations of equation renderers for websites. I eventually put together an algorithm that processed LaTeX equations through latex, dvips, and ghostscript to produce png images containing the equations. I also wrote some post-processing that would allow the equations to be properly aligned with text: . (I found the idea in one of the several rendering projects that I looked at, and improved on it.) My text processor for Movable Type, KwickXML, has included my equation rendering code from the beginning. I recently updated/rewrote KwickXML and have rewritten my equation rendering code simplifying and improving the algorithm.

I now just run the equation through pdflatex, generating a pdf file that I then post-process. I no longer use latex to add color to the equation, instead I do that in the post-processing. On rendering, pdflatex produces a white equation on a black background, which I use as a mask in post-processing. When paired with another image, the white on black mask determines how opaque the pixels in the first image are, producing smooth, alpha blending. (Note that the following example may not work on browsers that don’t support pngs with alpha transparency: *cough* IE6 *cough*.)

The Mask

Alpha Transparency

Alpha Transparency

Pixel Transparency

Pixel Transparency

The image with alpha transparency blend seamlessly with both background colors. The image with pixel transparency only looks good on the white background. Its nice anti-aliasing falls apart when it is placed on a color different than the one it was designed for—in this case white.

When using equations in webpages, it is difficult to align them properly with the text: . Out of the many options for vertical alignment in CSS, I decided to go with “vertical-align: middle”. However, not all equations display properly under such vertical alignment: a+e=c () and a+q=g ().

To properly align the equations with text on the webpage, I need to determine the “mean line” of the text and add space either above or below the rendered equation so that mean line is at the middle of the final image. To accomplish this, I add a dash before the equation, which will appear on the mean line:

I then process the image, using the dash to determine where the middle of the image should be and add any space as needed, remove the dash, and produce the final mask.

Putting it all together, we have the complete algorithm:

  • insert equation into a LaTeX template
  • render with pdflatex
  • trim image, removing extra blank space
  • use dash to detect the “mean line”
  • remove dash and trim again
  • add space above or below image to make the dash line in the middle of image
  • compose the mask onto the foreground color with the “CopyOpacity” option
  • compose that image over the background color
  • save image

Of Nests and Bands

Who knew that a group of wombats was a “wisdom” or a group of walruses was an “ugly” or a group of rattlesnakes was a “rhomba”?

Canadian Mind Products knew; check their site for more collective names and post the interesting ones.

About this Archive

This page is an archive of entries from October 2007 listed from newest to oldest.

September 2007 is the previous archive.

November 2007 is the next archive.

Find recent content on the main index or look in the archives to find all content.

Archives

Powered by Movable Type 4.37