Skip to main content
Move the World.
The Big Stories You Might Have Missed
MRI scan; brain cancer (glioblastoma) Credit: Wellcome Collection

Here are five big stories you might have missed last week:

1. Can Polio Cure Brain Tumors? (STAT News)

A form of poliovirus could help patients with an aggressive, incurable brain cancer known as glioblastoma, according to a new study in the New England Journal of Medicine. The virus in the polio vaccine is modified to target cancer cells and then injected directly into the tumor. The idea is to trigger an immune response, prompting the body to attack the infected tumor cells. 

A Phase I clinical trial showed that 21% of patients who got poliovirus therapy were alive three years later, compared to just 4% in historical data. That seems promising, but the study was small — only 8 actual patients made it to three years — and many small cancer studies show such "long tail" survival rates.

2. A Thousand Genes Are Linked to Intelligence (Science Magazine)

A massive new study has linked over 1,000 genes to intelligence. Using powerful algorithms and a database of 270,000 volunteers, scientists identified genes correlated with IQ and other health factors. Genes connected with intelligence carried a higher risk of autism, but they also appeared to protect against Alzheimer’s, ADHD, depression, and schizophrenia.

A separate study by the same researchers found nearly 600 genes correlated with neuroticism. Depression was associated with a distinct sub-group of genes from anxiety, suggesting that (although they often occur together) the conditions have different biological pathways.

3. Minting a Cheaper Nickel, Using Simulations (New Atlas)

It costs 7 cents to make a nickel, and that's just not good business. The U.S. Mint is looking to cut the cost of the metal to less than its face value, while maintaining its look and strength (and it still has to be recognizable to vending machines and coin-counters). But inventing and testing new metallurgical compounds by trial and error is expensive and time-consuming.

The Mint turned to the Materials Genome Initiative (an ambitious project to simulate how every known metal interacts) for help. Using MGI's data, they designed a computer model with specifications for strength, color, shine, and magnetism, and the program simulated a new alloy for nickels that meets every criteria but costs 40% less. Simulations like this could help industries rapidly invent new materials, without the cost of prototyping every design.

4. Concepts Expand When Examples of Them Fade (ScienceDaily)

A new study by Harvard psychologists shows that people's definition of a concept expands when examples of it become rarer. They call this tendency "prevalence-induced concept change," and, in several different experiments, they show that it can alter people's judgments about everything from colors to ethics.

Summarizing the results, they conclude, "When blue dots became rare, purple dots began to look blue; when threatening faces became rare, neutral faces began to appear threatening; and when unethical research proposals became rare, ambiguous research proposals began to seem unethical." (The study is published in Science; excerpts can be read here.)

5. 35,000 Studies Might Need to Be Retracted (Retraction Watch)

A new study suggests that 35,000 biomedical papers might need to be retracted because of duplicated or manipulated images — and ten times that number may need corrections or updates. It's a small fraction of the total scientific literature (and the numbers should be taken with a grain of salt), but it's a bit concerning that so many peer-reviewed journals could be missing problems like this.

By scanning over 20,000 published papers, researchers discovered that nearly 4% contained images (or parts of images) that were copied and reused with different labels — fudging the data, deliberately or not. A closer look at the duplicates from one highly ranked journal found that at least 10% needed to be totally retracted. Using those figures, they estimate that over 330,000 studies published since 2009 probably have altered or mislabeled images, and roughly 35,000 might warrant retraction.