I can’t be as confident about computer science as I can about biology. Biology easily has 500 years of exciting problems to work on. It’s at that level.
-Don Knuth on Biology
Not surprisingly, the recently announced “DARPA Mathematical Challenges” is full of biology related problems.
“DARPA seeks innovative proposals addressing these Mathematical Challenges. Proposals should offer high potential for major mathematical breakthroughs associated to one or more of these challenges.”
The specificity of the challenges listed is rather variable. Some are very clearly defined:
Mathematical Challenge Nineteen: Settle the Riemann Hypothesis
• The Holy Grail of number theory.
While others are extremely ambiguous and probably not resolvable just through mathematics.
Mathematical Challenge Fifteen: The Geometry of Genome Space
• What notion of distance is needed to incorporate biological utility?
Overall the list looks like the result of a fun brainstorming session and gives some focus to grant applications.
Created using Context Free on a Mac. Generative art is a staggeringly fun time waster.
From the Anabolic Halo web site (emphasis mine):
“Then, ANABOLIC HALO exploits insulin, one of the body’s most powerful musclebuilding hormones. This modulates the Anabolic Transcription Factor (ATF-4), which plays a critical role in the regulation of muscle anabolism and catapults you into a powerful anabolic state. ANABOLIC HALO directly manipulates insulin and regulates the activity of the key biochemical mediators of muscle catabolism, to simultaneously control the switches that restrict muscle size and create a rich anabolic environment primed for ridiculous muscle growth.”
Really, I don’t think there’s enough talk of transcription factors in advertising. This represents a new high in nutritional advertising pseudoscience.
It’s bad idea to name a system using an meaningful acronym, as the function of systems tend to drift over time, and over-specifying function in the name will likely render it inaccurate in the future.
Two good examples from biology are:
It’s really not a huge problem, but if you happen to be naming something you expect to persist for a while, it’s better to choose a name that doesn’t imply the purpose too explicitly as it’s likely to change over time. And it’s really hard to change a system name once it has been used for a while.
I don’t think the name of a system matters that much, but I don’t like them when they’re misleading.
The popular science press rediscovers “junk DNA” every few months, usually with an article breathlessly pointing to some discovery or other where it’s claimed DNA previously believed to be “junk” turns out to have a some function or other. Gosh!
You’ll be hard pressed to find a peer reviewed reference where they claim that any given stretch of DNA in a genome is “junk”. The term is virtually unused in scientific publications compared to the net as a whole (searches for “Junk DNA” in PubMed & Google give 78 and ~256K hits respectively).
The Wikipedia discussion on the junk DNA entry has comments recommending that it be merged with Noncoding DNA. I’m not sure that’s a good idea, as it’s not really synonymous. “Junk DNA” is an expression of an idea that most of our genomic DNA has no function, but this is rather misleading and far too teleological an idea to apply to the products of evolution.
“If the Vikings were around today, they would probably be amazed at how much glow-in-the-dark stuff we have, and how we take so much of it for granted.”
— Jack Handey [Deep Thoughts]
Buying some glasses online for my Dad I ran across these ones at optical4less. I’m very close to buying them. The fact that you can buy high quality glasses for $50 makes frivolous purchases a lot more likely: this also points to a new business model selling cheap glasses for all occasions. Work glasses, semi-formal wear glasses, glasses specifically optimized for eating at McDonalds (grease shedding!). Angry optometrists feeling threatened by online glasses purchases should take note!
Just for fun looked at the number of GEO (Gene Expression Omnibus) sample submissions per submitter, ranked by number of submissions. Plotting the top 400 with the X axis being rank and the Y axis number of submissions, I note that it follows a very nice clean power law distribution (R^2 = 0.9957)!
I’m sure I’ll go to bioinformatics hell for graphing this in Excel and not R or gnuplot, but there it is.
It’s amazing where power laws show up if you look for them. I suspect most of this is due to power law distributions in size (and activity) of organizations/labs submitting.
About a week ago I saw a City of Ottawa employee dowsing at the corner of Alta Vista and Main. He was walking across a lawn using a couple of bent metal dowsing rods. I’m guessing he was looking for the location of a buried water pipe.
Is this standard practice for municipal employees?
From The Onion, an old one, still funny.
New Scientist has a blurb on an interesting paper on correlations between neighborhod walkability and obesity: Living on the wrong street could be making you fat, however I can’t find it at the American Journal of Preventative Medicine (AJMP): I think the citation might be wrong.
There are however many good articles about the association between urban design and weight/health in AJPM: just search for “built environment” on the site. Even if you don’t have access the abstracts are quite accessible.
Scanning a few related papers shows that there is a strong correlation between the age of neighborhoods and weight as older neighborhoods are more walkable in general. I can see how that would be the case: even where newer neighborhoods have sidewalks there isn’t much to walk to. We chose our house largely based on walkability: I always want to live in a location which doesn’t require me to own a car and we walk quite a bit as a result.