Week in Review: Eye-Level

UNC-TV Science Week In Review: August 15, 2013


Grand theories, new organisms and brilliant cures often steal the show when it comes to common science knowledge. We can see medicines healing the sick, shark teeth the size of our fists and the tides rising and falling as the moon completes an orbit.

What can sometimes be forgotten is the research that goes into being able to detect changes. The tiny alterations in our DNA that change our appearance, the particular wavelength of light needed to vibrate water molecules and heat up food (microwaves), the tiny particles that can explain why we have mass. Somebody not only had to invent the tools to see this stuff, but also connect it back to what we can see.

This week, scientists from North Carolina took some of these invisible phenomena and brought them up to eye-level.

Changing Brains
Two major studies, one from Duke and one from UNC Chapel Hill, examined how minute changes in our DNA effect how our brains are constructed, and new ways to characterize and hopefully treat major mental disorders.

First, scientists from Duke collaborated with other members of Epi4, an international research group, to look for common genetic mutations in patients with epileptic encephelopathies (commonly known as EE’s or epilepsy).

EE’s start at a young age and cause debilitating seizures. The causes of EE’s are largely unknown, but a few cases have been shown to have a genetic link. The researchers compared the genomes of 264 EE patients and their families.

They found 329 mutations in total, 25 of which were shown to cause EE. Some of those 25 were found on genes not previously linked to EE, giving researchers new information on which genes could be targeted for early detection and treatment. The paper describing this work was published in Nature and the authors estimate this data describes the cause of more than 10% of EE’s.

Meanwhile, in Chapel Hill, another group identified common genetic traits of five major mental disorders: Bipolar Disorder, ADHD, Autism Spectrum Disorders, Schizophrenia and Major Depressive Disorder.

They used genomic data from more than 32,000 patients with these disorders and matched up common changes to a single letters in their DNA called single nucleotide polymorphisms (SNP’s). 

After combing through more than 4.5 million SNP’s, the researchers found that schizophrenia was 68% and 43% correlated to bipolar disorder and major depressive disorder respectively. Major depressive disorder was 47% and 32% correlated with bipolar and ADHD. Schizophrenia and autism were only correlated at 16%.

Their paper appeared in Nature Genetics.

Isotopes Identify Polluting Mines
Identifying mining pollution can be difficult. Why? Because the pollutants coming from most mines are generally made of the same stuff that comes from other sites: metals, gases, salts and minerals that are found naturally in the Earth, but can harm rivers and lakes. This difficulty spells trouble for regulators, who can find pollutants in the water, but can’t figure out whether they came from the mine, or someplace else.

That’s where professor Avner Vengosh from Duke University comes in. He and his team figured out a way to use the isotope composition of these pollutants (namely strontium, sulfur and carbon) to determine mine pollution from other sources.

Isotopes are forms of the same element with different numbers of neutrons. A common example of isotopes is carbon-12 and carbon-14, carbon-12 is more abundant and has 6 neutrons while carbon-14 is used in carbon dating and has 8 neutrons.

They found that the ratio of one isotope to another is different when they come from mountain top mining, so by measuring how the isotopes compare, they can determine whether the pollution was leached naturally from rocks or came from a mine. Their research appeared in the journal, Environmental Science and Technology.

“The Orc’s gold is second-most critical to the team.”
Any scientific paper containing the above quote is bound to be awesome.

Dr. David Roberts and Pu Yang of NC State University investigated how teamwork plays into online real-time strategy (RTS) videogames. Games like Starcraft, Warcraft and Defense of the Ancients rely on a team strategy to win. Players must share resources and control their growth so that the whole team can prosper.

Take those Orcs for example. If they gather gold too quickly, there won’t be enough for their teammates to prosper. On the other hand, if they don’t gather it quickly enough, they will be too weak to help their team in a battle. It’s a very delicate balance.

What Yang and Roberts did was read the logs from thousands of games and design an algorithm to describe a team’s chance of winning a game based on how quickly their characters grow.

They presented their results both in a fascinating paper (that you really don’t need to know videogames to enjoy) and at the IEEE Conference on Computational Intelligence in Niagara Falls, along with another paper. Their findings will be useful both to game designers and to those who study decision making in team-based activities.

- Daniel Lane

Daniel Lane covers science, medicine and the environment as a reporter/writer. He is currently pursuing a master's degree in medical and science journalism at UNC - Chapel Hill.