Oops. Correcting scientific errors

A recent study reported finding deadly germs on a subway — except they weren’t really there

microbes NYC subway

This is a map of the microbes on the New York City subway system. But uh oh: Two of the very dangerous diseases that scientists had said were there really weren’t. 


Did you hear there were bacteria that could cause bubonic plague and anthrax on the New York City subway? Many media outlets reported that scientists swabbed the subway and found genes from the bacteria that cause these deadly diseases.

It now appears those scientists were wrong.

There probably were no deadly germs on the subway — certainly none that riders need to worry about. The scientists who wrote the paper have just admitted that. And they did so by issuing a correction — a new statement in the scientific journal in which they had reported their original findings.

Such a formal correction will describe one or more factual errors — or mistakes in how the authors had interpreted their data. In this case, the journal also changed the original paper, noting that one or more of its details have been corrected.

Corrections are always embarrassing to the authors. But they also are essential to scientific research.

Finding underground germs

“[Corrections are used] when scientists themselves realize they made a mistake,” explains Daniele Fanelli. He’s a scientist at Stanford University in Palo Alto, Calif., who studies how research is done and communicated. “Human beings are naturally prone to error, and scientists are human beings,” he says. So as mistakes occur, “you take care to correct them.”

In this case, researchers had mapped the metagenome (Met-uh-GEE-nohm) of the entire New York City subway system. Basically, they cataloged all of the genetic debris left behind by organisms that had come through the stations. To do this, the scientists — led by Christopher Mason at Weill Cornell Medical College in New York City — sampled the floors, walls and more at all 466 subway stations.

Each animal, plant, bacteria and fungus has its own unique genes. Scientists often can confirm the presence of a specific species just by studying the genetic residues it left in its wake. And Mason’s team found the subway was full of genetic castoffs. Many came from animals, including people, rats and insects. But the genetic debris also showed the subways were full of microbes. Most are harmless. But the genes for a few seemed potentially worrisome.

For instance, Mason’s team reported finding genes that appeared to come from the bacteria that cause anthrax and the bubonic plague — two very deadly diseases. They also found genetic material from germs that appeared to be resistant to the drugs doctors typically prescribe to kill them.

The scientists published their first report on March 3 in Cell Systems. “Within two or three hours of the paper coming online, people were poring over the data,” Mason recalls. Journalists and bloggers started writing about New York’s subways hosting plague and anthrax germs.

But some scientists became concerned about those claims, Mason says — and for a different reason. In examining the paper’s published data, they found that the evidence of germs for anthrax and plague was quite weak at best.

Soon, scientists at the New York City Department of Health and the U.S. Centers for Disease Control and Prevention wrote a response letter to the scientific journal. In it, they suggested the data pointing to deadly bacteria might not be solid. The scientists at the New York City Department of Health and the CDC also became worried by the media coverage the paper had been receiving. As people read about plague and anthrax germs, they might needlessly panic about disease risks. Plague, for example, is not normally found in New York City. And no one in the city was sick with anthrax.

Mason and his group read the health department’s letter and decided to double-check the claims in their paper. To do this, they performed more experiments. They used the same genetic material they had gathered from the subway. But this time they probed for different genes associated with anthrax and plague bacteria. In particular, they looked for signs that would mark virulence — the ability to cause disease.

And this time, Mason says, his group found no sign the bacteria were virulent. This made it very unlikely that the genes his team had picked up came from either anthrax or plague bacteria. The scientists also retested their samples for signs a microbe had been drug resistant. Again, this time, no signs of resistance emerged.

Correcting the record

With these new findings in hand, Mason and his group sent a formal scientific correction to the journal that had published their initial paper. This shorter note identified what details or phrases in their previous paper were wrong. It also described the new follow-up experiments they had performed — and how these contradicted their earlier conclusions.

A correction is not a retraction. That’s a formal report stating that the authors of some published research paper are withdrawing that paper. Doing this informs the research world that the scientists no longer stand behind what they had published earlier.

In the case of Mason and his group, they still believe their main microbe map of the New York City subway is valid. They also believe their results still offer a useful catalog of the many types of organisms that travel through the subway system. Their correction just fine tunes details of their work to make a few parts of it more accurate. 

Asking a journal to run a correction is always a hard thing to do. “No one likes admitting error,” says Ivan Oransky. He is one of the founders of the website Retraction Watch, where he writes about scientists who make mistakes — and some who commit deliberate fraud. Oransky notes that he makes mistakes, too. And it’s never pleasant when others point them out. “When a reader points out an error, my first response isn’t to jump up and down with glee and say ‘Great! I made another mistake!’”

After acknowledging that you made a mistake, Oransky says, “your first thought is that people will trust you less.” But he notes that instead, when scientists state they’ve made a mistake, and then correct the record, other scientists actually trust them more.

Corrections are an example of how scientists make progress toward a better understanding of the world. After all, most new findings are based on previous scientific research. “Every scientist or team of scientists is adding a little brick on a complex and ever-increasing building,” explains Fanelli. “It is essential that each brick stands on other bricks that are sturdy and reliable.” New evidence may prove old evidence wrong. But learning that is important. And, he adds, it helps science move forward more strongly than ever.  

Follow Eureka! Lab on Twitter

Power Words

(for more about Power Words, click here)

anthrax  A bacterial infection of sheep and cattle that also can be transmitted to humans.

bubonic plague  A disease caused by the bacteriumYersinia pestis. It’s transmitted by the bite of a flea that had previously bitten some rodent infected with the germ. This form of plague causes fever, vomiting and diarrhea. It also inflames the lymph nodes, causing them to swell. Those swollen tissues, called buboes, give this form of the disease its name. Known as the Black Death, bubonic plague killed millions of people in Europe during a series of outbreaks during the Middle Ages.

correction  An addition issued to a scientific paper by the researchers, stating they have made errors and correcting any incorrect statements.

gene   (adj. genetic) A segment of DNA that codes, or holds instructions, for producing a protein. Offspring inherit genes from their parents. Genes influence how an organism looks and behaves.

genome    The complete set of genes or genetic material in a cell or an organism. The study of this genetic inheritance housed within cells is known as genomics.

genetic  Having to do with chromosomes, DNA and the genes contained within DNA. The field of science dealing with these biological instructions is known as genetics. People who work in this field are geneticists.

metagenome  The genetic material recovered from samples in the environment. Studies of the genes recovered from stream samples, soil and even swabs on a subway car can reveal the plants, animals and many bacteria that might have recently been present.

microbe  Short for microorganism. (see microorganism)

microorganism  A living thing that is too small to see with the unaided eye, including bacteria, some fungi and many other organisms such as amoebas. Most consist of a single cell.

resistance    (as in drug resistance) The reduction in the effectiveness of a drug to cure a disease, usually a microbial infection. (as in disease resistance) The ability of an organism to fight off disease. (as in exercise)  A type of rather sedentary exercise that relies on the contraction of muscles to build strength in localized tissues.

retraction    A formal announcement that researchers (or the organization that may have published their findings) no longer stands behind the published data. The initial report of data will not disappear. It will just be flagged with a warning that the authors or publisher no longer trusts the data or findings as reliable.

virulence  (in medicine) The potency of a virus, bacterium or other agent in causing infectious disease. Among a given species, some strains may cause disease with very little exposure (such as infection with a few cells). Less virulent strains may take massive exposures to create disease.

Yersinia pestis  The bacterium that causes plague, both the bubonic and pneumonic forms. 

Bethany Brookshire was a longtime staff writer at Science News Explores and is the author of the book Pests: How Humans Create Animal Villains. She has a Ph.D. in physiology and pharmacology and likes to write about neuroscience, biology, climate and more. She thinks Porgs are an invasive species.