Controversial ‘arsenic life’ paper retracted from Science

After almost 15 years, the influential scholarly journal Science has retracted a controversial paper about the discovery of a microbe able to use arsenic instead of phosphorous in its biochemical processes (Science 2010, DOI: 10.1126/science.1197258).

In the retraction notice, H. Holden Thorp, Science’s editor in chief since 2019, states that the paper is being retracted on the basis that the “reported experiments do not support its key conclusions” but not because of any deliberate fraud or misconduct.

The retraction is accompanied by a letter from the authors disagreeing with the decision, stating, “While our work could have been written and discussed more carefully, we stand by the data as reported.”

In the original paper, the authors describe an extremophile microbe called GFAJ-1 that they believe has the ability to weave arsenic into its proteins and nucleic acids to compensate for a lack of phosphorus. Confirming that arsenic can be incorporated into DNA has the potential to expand the chemical understanding of life on Earth—and beyond, an aspect that NASA and the journal initially played up.

The paper immediately garnered harsh criticism from the scientific community over the researchers’ methods and conclusions. In 2012, Science published two papers from independent teams who were unable to replicate the original paper’s results.

The current consensus is that although GFAJ-1 is unusually good at growing in high concentrations of toxic arsenic, sufficient evidence hasn’t been found that it actually incorporates arsenic into its DNA, and the bacterium likely survives by scavenging trace phosphorus from its surroundings. The authors say in their letter that these follow-ups didn’t adequately reproduce their original growth conditions for GFAJ-1.

Back in 2012, Science’s policy was to retract papers only in cases of misconduct, which this is not. But, Thorp says, “The expectations for straightening out the literature have risen significantly,” and calls for the paper to be retracted have simmered throughout his tenure at the journal.

The tipping point was a New York Times profile of Felisa Wolfe-Simon, the paper’s outspoken first author, that was published in February 2025. Thorp went on the record in that article saying he thought the paper should be retracted.

According to guidelines from the Committee on Publication Ethics (COPE), “Retraction is a mechanism for correcting the literature and alerting readers to articles that contain such seriously flawed or erroneous content or data that their findings and conclusions cannot be relied upon. Unreliable content or data may result from honest error, naïve mistakes, or research misconduct.”

Science’s current retraction policy states, “An accumulation of errors identified in a paper may cause the editors to lose confidence in the integrity of the data presentation, and the paper may be retracted.”

Ariel Anbar of Arizona State University, one of the original paper’s authors, says the uncertainty in the data is “larger than one would like,” but “we don’t think there was a major error” that warrants retraction.

Anbar says Thorp asked Wolfe-Simon to voluntarily retract the paper last fall, after he was contacted by the New York Times. She refused. After months of negotiations, the researchers and journal editors reached a compromise: The paper would be retracted with language that both parties agreed on. The researchers could voice their dissent in a letter.

In addition to posting the retraction notice, Thorp and Science’s executive editor, Valda Vinson, published a blog post in which they expand on their reasons for retracting the paper. In the post, they reiterate that the decision is rooted in experimental error and “at no point has there been any discussion or suggestion at Science of research misconduct or fraud by any of the authors.”

Thorp says he hopes that retracting the paper will put a period on the whole affair. “I hope this is the end of it,” he says.

Anbar says he and his coauthors asked to see the blog post ahead of time, and the journal didn’t respond; they received a copy from a reporter. The post brings up critiques beyond what went into the notice, which Anbar says undermines the good-faith commitment to transparency that the authors and editors made when negotiating the retraction.

One thing both Anbar and Thorp appear to agree on—although for different reasons—is wishing that arsenic life hadn’t whipped up such a firestorm in the first place.