First, I’ll say that I actually like the little stand-alone SFS program: I find it quite user friendly and versatile. But a version accessible via the web is nice (and inevitable). The online version has kept most of the same interface and function, with a web style.
Other databases, notably ISI Web of Science, are really great at finding H-index values or citation trends. But, when I really want to do a lit search on a topic or reaction, SFS is always my first choice (maybe after just typing it into Google). CAS has a nice Flash demonstration and screen shots here. Here are some on my own screen shots (click on them for full images).
A fruitless search:
And the result. Or a reaction search:
Another cool new feature in SFS (web or software version), is the “keep me posted” button: You can follow up on searches or results. Currently, I have alerts from ISI sent to my RSS reader; I’ll have to explore this new SFS feature to see if it’s better.
I think I’ll use SFS web, especially when I’m using a computer that doesn’t have the software installed. Overall rating: super rad cool.
My ACS feeds sometimes screw up and send me RSS articles for the wrong journal. Does this happen to anybody else? Sometimes I don’t notice for several articles that I’m reading titles from Inorganic Chemistry instead of JACS.
Sometimes it’s more obvious:
I was pretty impressed at first, when I thought someone got a paper on Killer Whales into JPCA!
[Update: ACS claims to have fixed their feeds. I haven’t had a problem since.]
I’m not a huge fan of the Gaussian online manual. Maybe I’m spoiled by Matlab, but G03MANTOP is soooooo confusing and not user-friendly. Even William thinks it’s tough to understand. Here’s my favorite line I’ve found:
CIS Output. There are no special features or pitfalls with CI-Singles input.
Maybe this isn’t funny to everyone, but I just like the idea that they didn’t add any special pitfalls to the CIS input format.
(BTW, that claim is bull. I haven’t been able to get my CIS calcs to work too good so far.)
My lab has a few Andor CCD cameras for imaging, and I was happy to find an plugin for opening Andor .sif files directly into ImageJ.* I was happier to see that my friend Stirling Churchman wrote the plugin originally. Thanks Stirling!
* ImageJ is a great open-source image processing program developed by the NIH. I’ve been really happy with it (especially the plugins!).
I need some help with my Matlab m-file. Any geniuses know how to save a figure without the border?
- The m-file produces an array of numbers: columns indicate the x- and rows the y-position of a pixel of intensity denoted by the number’s value
- I make a figure and use the command
- I turn the
- I save using the
Basically, I want the
pcolor figure alone so I can import it into a different program and click on the image to move a stage to that location. Having a border means the stage will move to the wrong place. (I know this is a clunky solution to a scanning-stage program. Shut up. Stop making fun of me.)
Any thoughts? I’ve found that
imwrite saves only the image array, but I don’t know how to use
shading 'interp' with
To encourage random people to help me with my research, I’ll send an EDS magnet to the first person who tells me how to do this correctly—or at least points me in the right direction. I’ll update y’all as soon as I know.
[Update: Here’s a good image I created. See comments below.
I know, it’s flipped. But I can easily deal with that.
We have a winner, but I might send more magnets for more elegant solutions.]
Proven: Using the internet makes you live longer. And, as a nice side effect, the internet reduces child mortality. Don’t believe me? See for yourself (click on the chart below):
(To manipulate the chart or see a map or make your own chart, click here.) Cool Google stuff.
The wave of social bookmarking is finally reaching scientific journal articles. What YouTube, de.licio.us, and digg are to stupid videos and strange websites, now biowizard, CiteULike, and BioInfoBank are to scientific articles. (See also my previous post about CiteULike.) I suspect Connotea, with some social features already existing, is on its way to more filtering of scientific articles, too.
I welcome this progress. I find it pretty tedious reading journal tables of contents every day/week, which is necessary to keep up with new ideas that you hadn’t thought of (and thus are hard to search for in a database). Also, with journals such as PLoS One (which publishes anything—whether it’s interesting or not), filtering of journal articles will become more and more useful to pinpoint only articles you’re interested in reading.
The ranking mechanism of these science social bookmarking sites is far from perfect. But it’s a good start. Go check them out and make it easier for me to find papers I want to read!
[Note: My favorite site is still CiteULike. Connotea is nice and probably less buggy, but it seems to progress at a much slower rate, probably because CUL programming is open-source and collaborative. I can’t imagine biowizard really being that helpful in the long run, because it doesn’t have any of the functionality of CUL or Connotea and only applies to biology. I’m going to stick with CUL until someone else wins unequivocally (then I’ll export my library). I’ll update you as things change.]
Don’t you sometimes regret that the little search field in the upper right corner of a Firefox window (or the Google toolbar or … I dunno, does the new IE have a search field?) remembers the stupid things you type in? Is there a word for search-memory regret?
So embarrassing. Especially the misspelling. (By the way, do not search what I searched. Very disturbing.)
A better search history here.
I recently found a cool website: CiteULike. It allows you to easily post scientific articles that you find interesting. To post an article you find, you simply stay on the page with the article (or abstract) and click a CiteULike bookmark, which automatically imports the various fields (e.g. authors, title, journal, etc.) into your library; then you simply add your own “tags” (or categories) and submit the post. I love it!
This site has a two-fold benefit: (1) you can share this library with others who are in your field (see my Watchlist), and (2) you can rank them according to how much you want to read them and remember to read papers.
CiteULike was written by Richard Cameron in 2004 and is run by him. And it’s free! I was so happy to find CiteULike, because I came up with the idea a few weeks ago, but I don’t have enough programming experience to write such a website (I took a C++ in high school, but didn’t really do the homework).
There are probably several other sites like this one, but the only other one I’ve found is Connotea, which was commissioned and now part of Nature Publishing Group. I checked out Connotea, and even imported my CiteULike library (tags include). But I prefer CiteULike’s style and features, especially the ability to rate and sort how interested I am in reading each article (maybe that feature exists in Connotea, but I can’t find it). I hope that CiteULike eventually becomes an open-source, collaborative effort; but until then, I’m still really happy with it.
So start you own library and send me a link!
My brother showed me this cool Applet from NASA: you can track tons of satellites! My favorites are the many satellites in geosynchronous orbit—this is called the “Clarke belt” after the author Arthur C. Clark, who discovered/promoted the idea of placing satellites in an orbit with radius 42,000 km. Also, if you click on “Satellite -> Select” and choose “Human Crew” under type, you can see what I presume is the International Space Station. But where’s the type with “Simian Crew,” so we can see how the monkeys are doing?
I’ve been doing some electronic structure calculations recently, and this is a job for Gaussian03. Sweet Gaussian. It’s a fine choice for electronic structure calculations, but there are some true oddities.
1. The program is probably the most popular suite for solving quantum mechanical calculations. Thousands of chemists use it and pay through the nose to do so. However, it is almost completely undocumented. For example, see the description of the transition option under the Density keyword (Figure 1). What is the transition density? Ought I know? Basically, do any calculation with any method other than B3LYP/6-31G and you’re screwed.
Transition=N or (N,M)
Use the CIS transition density between state M and state N. M defaults to 0, which corresponds to the ground state.
Figure 1. Out of context? no. See for yourself.
2. Why does it crash? I realize it’s a complicated system and lots can go wrong. But, why doesn’t the output file give any useful information about the system failure. It’s a smart program. It should know why it failed and speak its mind. Figure 2 shows one of my favorite error messages. Of course, there is no content about such errors in the documentation.
Erroneous write during file extend. write 172031 instead of 4096
Write error in NtrExt1
Figure 2. Actually, I know what this error is. The computer ran out of hard disk space. But, it wasn’t easy to figure out. Try googling the error message. You’ll find a bunch of people trying to figure out what NtrExt1 means.
Why isn’t there a wiki-gaussian? I don’t know, but I hope I’m not banned for asking.
The folding-at-home package has been ported to Sony’s much anticipated PS3 console. The point? None! But, it’ll make Vijay’s group happy and its always nice to see some of our peers get street-cred in the popular media. I hear their groundbreaking cell architecture is a ~10x improvement over your typical PC!
Ah, the irony – this is probably the only software out there that can leverage the PS3’s sweet sweet cpu cycles. Now, if I could only frag more of those proteins … sigh … here is some eye candy:
How do you turn your grayscale CCD to a two-color camera? Filters and fun! Here’s a diagram looking down at the dual-viewer setup on my table:
Figure 1. Diagram of setup (viewed from above)
M1 and M2 are mirrors, F1 and F2 are long- or short-pass filters, and DC1 and DC2 are identical dichroics. DC1 reflects short wavelengths and pass long wavelengths, the filters clean up the two paths, the mirrors bring the paths back together, and DC2 combines the two channels. If the channels are offset a little, then short and long wavelengths are split into two copies of the image onto the CCD.
Here’s a pic of the setup:
Figure 2. Picture of the setup with channels drawn
In other words, the dichroics split the green light off the output and move it to a different region of the CCD. You can recombine the two copies and add color using ImageJ, like this:
Figure 3. The right side is an overlay of the red and green channels in false color
Now, there’s probably already a program out there that will compare the optimized structures of quantum-chemistry calculations (e.g. Gaussian). But I couldn’t find one, and I wanted control over how to compare the structures (for instance, I wanted to be able to ignore the errors in the dihedrals for light atoms). So I wrote a MatLab M-file that outputs the absolute differences in the bond lengths, angles, and dihedrals between two structures. It took me a while to get what I wanted out of Molden and make MatLab spit out the right set of numbers, but I finally got it to do what I wanted (I think).
It’s not perfect, I’m pretty happy with it. It just imports the variables from a Molden z-matrix and takes the absolute difference in them, sorting my type. Then I just throw those outputs into SigmaPlot or Excel (ew!) or whatever and do the error analysis I want. For instance, I remove the dihedrals that don’t really matter (some light atoms and some artificially inflated dihedral errors) and just look at the mean and standard deviation for different method/basis set combinations.
I dunno, not essential. But I find it pretty handy. If you want to check it out or try it, I’ll post it below. Let me know if you find errors or if you find ways to make it better.
Dealing with engrained technology can be a little frustrating. Many years ago, a grad student wrote a program in LabView to fit Gaussians and exponentials. This is a simple problem, and there are many fine programs for dealing with numerical analysis. The problem with LabView is that it isn’t one of them. As shown in Figure 1, it gets a little complicated.
To be fair, it does work. It fits mighty nicely after it is debugged and running, as shown in Figure 2. As my co-worker says, “When it works it works well.”
Figure 2. A good fit.