Map of melting in Greenland published by NASA. Credit: NASA/Earth Observatory.

NASA is reporting of new data about the melting in Greenland in 2007. In short: areas at high altitudes (over 1.2 miles above sea level) are experiencing record melting, clocking in at a chilling 150% above average. At lower altitudes, the melting is 30% above average. To put this in perspective, the amount of snow that has melted this year over Greenland could cover the U.S. more than twice over.

The melting data came from satellite-based microwave imagery, which was then compared to the average snow melting from 1988-2006. The map on the left shows the melting difference between the past 19 years and this year: the redder it is, the more melting that occured in that area this year. All in all, this puts 2007 as the fifth highest in amount of melting. The top four are, in order: 2005, 2002, 1998 and 2004.

Needless to say, this is scary proof that global warming is really happening and that predicting and modelling how melting occurs is very difficult. And then we have to think about where this melted snow is going, i.e., to the seas, which brings up the thorny issue of rising sea levels. The image above is no longer pretty, is it?

I just came across the UK government's Teachernet's guidance on teaching creationism in UK schools. What's in this document? It's the best write-up I've seen about the evolution debate, along with comment on its place in science teaching in the UK. There is a list of questions, a glossary of terms, a thorough introduction to the topic, and then the guidance. It's written clearly enough for anyone to understand the question and the answer. In short, on teaching ID and creationism in science, the guidance says "don't teach it". Why can't everything be so well explained and clear cut? I love it!

Unfortunately it's not dated, so I can't figure out how new this is. The Word file stats suggest it was created September 2007 (File->Properties then select the Statistics tab). Regardless, it's a very useful document to keep handy for those inevitable times when this discussion comes up.

Some puzzles and brain teasers are not simple exercises - they're bordering on torture. But that's what makes a good puzzle good: it seems impossible and gets you all addicted till you solve it. So now that the days are getting shorter and the weather getting colder (at least for us folks in the Northern Hemisphere...), here are some logic and brain teasers to pass keep you company.

• wu: riddles: the more mainstream kind of logic problems classified as easy, medium, and hard. Prepare to waste hours on this collection.
• Classic Computer Science Puzzles from Coding Horror. If mathematical algorithms are your thing, these should be easy. Ha ha.
• Project Euler is a collection of math problems that require programming to actually solve.
• Finally, the shameless plug: blogSci's very own Daily Mental Calisthenics gives you word of the day, a Stroop effect puzzle and a Sudoku to kick start your day.

Please let me know your favorite collection of puzzles and brain teasers either in the comments below or by email.

Starchy diets affected gene copy number.

A new paper published last week looks at diet and the evolution of the human amylase gene copy number. The paper very nicely stacks a pile of evidence that starts to tell another story in human evolution.

Let's start at the top. In humans (and many other animals), the digestion of starch starts in the mouth. The enzyme responsible for starch digestion is called amylase, and specifically, the human salivary α-amylase. Given this genetic component, can we find variation between different human populations? The answer it turns out is yes, and it's correlated with the historical consumption of starch by these different groups of people.

The paper looked at three populations of people that have historically had a diet high in starch and four populations that have historically a diet low in starch. The data they uncovered was that people from high-starch populations tended to have more copies of the amylase genes - in one case 14 copies! - than people from from low-starch populations.

The conclusion that it's definitely evolution still needs more evidence, but the story is starting off to a great start. Why is the conclusion not 100% done and dusted? For each of the populations, we need to check two things:

• Is this variation found only in the amylase gene or is this a typical feature in the populations in question? To answer that, we check variation in other genes in these people and see what's typical and compare the amylase gene variation to the typical rate. The researchers of the paper couldn't do that for all populations, but for one population, a Japanese one, they did find that the amylase variation was significantly more than the most of the rest of the genome, i.e., something is going on and it's not random.
• Which copy number is the base case? One hypothesis that we need to check is that high copy numbers of amylase is the typical state in humans and that populations lost copies when their diets shifted to lower starch content. To fix this, we need to check more people from all over the world to figure out the details.

Still, it's a great first chapter of this story. Watch this space!

BPR3.org has just put up the three finalists of the logo contest for community voting. We last talked about BPR3 in a post thinking out loud about what such an icon should do. That post also has tons of references about the start of BPR3 and what it stands for.

So come on, get voting!

In the past week I came across two new online community sites, or to use the Web 2.0 jargon, "social networking" sites. One is for doctors and one is for patients.

The first one is sermo for doctors. The about page describes it as this:

Here, physicians aggregate observations from their daily practice and then - rapidly and in large numbers - challenge or corroborate each others opinions, accelerating the emergence of trends and new insights on medications, devices and treatments. You can then apply the collective knowledge to achieve better outcomes for your patients.

This is a very interesting idea. It's open to US-based physicians (which they check, according to the FAQ) and works by doctors asking and answering questions. The answers are voted on, and if you don't like an answer, you can add one. Very interesting.

Then you get to the 'how do we make money' part. This is how they describe it:

Sermo's business model is one of information arbitrage, the opportunity that arises when breaking medical insights intersect with the demand for actionable, market-changing events in healthcare.

Eh? Whoever wrote that needs get an award for Buzzword Compliance. How does it actually happen? It's better explained here. Basically, health care companies, investors, and consultants get a chance to ask the community questions. They pay sermo for the access, and whoever answers these questions may get paid for that.

That in itself is a very interesting business model (I've never seen anything quite like it), so if it works, Sermo could, gasp!, be a Web 2.0 company that actually makes money. We'll see.

Sermo has gotten some good coverage lately: the AMA has partnered with them and the WSJ had a piece about social networking that talked about them too.

On the other side of the coin is patientslikeme. They describe themselves like this:

Our goal is to enable people to share information that can improve the lives of patients diagnosed with life-changing diseases. To make this happen, we've created a platform for collecting and sharing real world, outcome-based patient data (patientslikeme.com) and are establishing data-sharing partnerships with doctors, pharmaceutical and medical device companies, research organizations, and non-profits.

This sounds like Sermo, but focusing on patients. There is one problem I see with this: although each page on patientslikeme clearly states that its content should not be taken as medical advice, I can't help but think that patients will go down the self-diagnosis route and take the content as medical truth, without regard if it applies to them or not. This problem is partly offset by focusing on a few diseases; at the moment these are ALS, Multiple Sclerosis, and Parkinson's disease. Still, this quote jumped at me from FierceHealthIT's award announcement for patientslikeme:

the site allows patients to select for other patients with a similar profile (say, same age, stage of disease, gender), see what treatments that patient has had and what their outcomes were.

Apart from potentially having patients 'suggesting' treatments to doctors, the site sounds like an amazing opportunity for patients and doctors to interact. To put it mildly, a world-wide support network for patients to help them through tough times is an idea whose time has come.

And there you have it. Two great online innovators in the world of medicine.