Consecutive Journal Publications Illuminate Collaboration And Compromise In Chemistry

Last April, Gary A. Molander got an unexpected e-mail. It had been a good spring—the University of Pennsylvania chemist had submitted a paper to the journal Science about a reaction that could broaden the scope of perennially popular cross-coupling reactions. With the paper out the door, he’d started discussing the work when visiting other departments. Word of the reaction had reached David W. C. MacMillan at Princeton University. The e-mail was from MacMillan, who wanted Molander to know that he and Abigail Doyle were working on something similar.

“Dave may have been afraid we were doing exactly the same thing, which could’ve been a disaster,” Molander says. The Penn team’s paper was almost through Science’s review process. That situation could have presented Doyle and MacMillan with an awkward choice: publish their work in a lower profile journal, or gamble with submission to a journal with a quicker turnaround time than Science, and scoop Molander.

The work wasn’t identical. Although both teams had discovered mild cross-coupling conditions for sp³-hybridized substrates, they were using the chemistry to make different types of products, in effect bolstering the case that the new cross-coupling was highly versatile. The teams revealed to each other that they were both submitting to Science, with Molander six weeks ahead. It was a quandary with a possible solution: publishing the work in two consecutive Science papers.

Chemists have been publishing back-to-back papers for as long as scholarly journals have existed. When the reports come from independent teams, their backstories shed light on collaboration, compromise, and competition—the human side of science.

Independent teams arrive at similar research because “people read the same literature,” says P. Andrew Evans, an organometallic chemist at Ontario’s Queen’s University who has published work back-to-back with another team on two occasions. “You have people formulating ideas in the same ways, and voilà, you get two groups with very similar papers.”

Molander sent MacMillan and Doyle a copy of his manuscript. The Princeton teams, who began collaborating on a single paper after learning about each other’s work at a very early stage, didn’t read it in detail to avoid confirmation bias when writing about their own results. They then sent their manuscript to Molander, who filed it away without reading it, so that the Princeton work wouldn’t influence his approach to revisions. When Doyle and MacMillan submitted their work to Science on May 1, their manuscript, like Molander’s, arrived at the desk of Jake S. Yeston.

A deputy editor at Science who specializes in chemistry, Yeston says he encounters a potential back-to-back paper scenario roughly twice a year. Science has no hard-and-fast rules that determine when two manuscripts should be published consecutively. Because of the breadth of research the journal must cover, Science rarely publishes similar papers a month or two apart. “We’d just about made the decision to go ahead with Gary’s paper by the time Dave and Abby’s paper came in,” Yeston recalls. Yeston didn’t yet know that the teams had spoken to one another. He didn’t think it was appropriate to delay Molander’s publication, so he contacted MacMillan and suggested he and Doyle send the paper elsewhere, for reasons he declined to disclose to MacMillan.

The exact number of back-to-back publications from independent teams is difficult to quantify. It’s easier to classify the situations that can lead to them. Sometimes, consecutive publications come from independent teams that reach completely different conclusions. More often, they occur when results are complementary. In these cases, independent teams may learn about each other’s work and coordinate submission to a journal. Or teams may be unaware of each other’s efforts, but their complementary manuscripts end up at the same journal by coincidence.

Author-coordinated cases develop for many reasons. Researchers may find out about each other’s work at a conference, or by simply keeping in touch with colleagues and mentors. In 1979, the latter was the case for Harvard University organic chemist E. J. Corey and his former student Hisashi Yamamoto, who was then at the University of Hawaii, Honolulu. The pair arranged back-to-back publication in the Journal of the American Chemical Society for their total syntheses of a metabolite from a marine sponge, 9-isocyanopupukeanane (1979, DOI: 10.1021/ja00500a048 and 10.1021/ja00500a049).

“In chemistry, the grapevine is pretty small,” Scripps La Jolla biochemist Richard A. Lerner says. In 1986, he and Peter G. Schultz, who was then at the University of California, Berkeley, coordinated submission of their work on catalytic antibodies to Science (DOI: 10.1126/science.3787261 and 10.1126/science.3787262). Lerner and Schultz were working in the same area but didn’t have an adviser-student relationship. “I don’t remember if I called Pete or he called me,” Lerner says, but he believes the decision to coordinate was the right one. “Rather than have an unfriendly situation for 20 years, we’ve been friends,” Lerner says. “It’s a better way to live.”

The Princeton-Penn case fell somewhere in between coordination and coincidence. “We didn’t know Gary was further along when we reached out to him,” Doyle says. Like Lerner and Schultz, she and MacMillan decided on openness. “We didn’t want to abuse the fact that we had privileged information” about what the Penn team was doing, MacMillan adds.

Scientists often come across privileged information at talks or as reviewers of grants and manuscripts. But they may choose to do different things with that information. “It’s not inconceivable to have a referee who deliberately delays a review, and you learn when the review finally comes in that the person’s own paper is in press elsewhere,” Yeston says. “That’s one of the challenges you worry about as a journal editor—inadvertently alerting rivals during the review process. But my general experience is that chemists are pretty honest people.”

Yeston’s cryptic suggestion prompted MacMillan and Doyle to contact him to discuss the situation. “It’s delicate when I have conversations like this,” Yeston says. His loyalty as a journal editor must be to the authors of the first manuscript submitted, he explains. When MacMillan and Doyle explained that Molander was amenable to back-to-back publication, Yeston couldn’t simply take their word for it—he had to contact Molander to confirm.

“If you have two similar papers come in, that’s cause for excitement,” says Jillian M. Buriak, editor-in-chief of Chemistry of Materials, an American Chemical Society journal. “If they’re published together, it inevitably draws more attention than the papers would get individually,” she adds. Buriak has yet to handle a case of back-to-back publication. She points out that numerous things have to fall into place for cases to work out. Because human beings are involved, she adds, the process is somewhat random.

If journal submissions aren’t coordinated by the researchers, “the two papers have to arrive close enough in time that someone at the journal connects them,” Buriak says. “If papers get sent to different associate editors, the connection may be lost,” she says. Whether submissions are coordinated, peer review timing is a big unknown. “Some papers require more revision than others, and the author whose paper was accepted more quickly may not want to wait,” she says. “There’s a personality aspect, a human aspect here that doesn’t often get publicized,” she says.

Back-to-back papers have been few and far between at Nature Chemistry, says Stuart J. Cantrill, chief editor of the journal. When complementary papers do arrive, his team strives for consistency in how the papers are evaluated. “You don’t want different sets of people reviewing similar papers” if you can help it, Cantrill says. “Being aware that a second paper exists can help put the first paper in context for a referee.”

Cantrill notes that even when submission is coordinated by two teams, a journal isn’t under any obligation to make the same decision on both papers. “Each paper should be evaluated on its own merits,” Cantrill says. If one paper didn’t review well or got bogged down with a lot of revisions he says, “I’m not going to hold the other paper for six months.”

Yeston reached Molander and asked him whether he would be willing to hold his paper in order for the Princeton team to catch up. It could take time, Yeston cautioned, depending on the outcome of the review process. Molander agreed.

“Jake was looking out for me,” Molander explains. The Princeton paper had arrived six weeks behind the Penn team’s—which could be construed as a long time in terms of journal submission dates. But in Molander’s view, six weeks was not that long a time compared with the five-year time to degree for the graduate students on the two papers who’d done the research in the first place.

When Molander put himself in the shoes of the students, “it showed that everyone’s priorities were in the right place,” Yeston says. “That’s when I felt comfortable that everyone was on board.” Doyle and MacMillan’s paper made it through the review process, and the papers appeared online simultaneously in Science last June and then consecutively in print (2014, DOI: 10.1126/science.1253647 and 10.1126/science.1255525).

Doyle’s and MacMillan’s teams continue to keep abreast of each other’s progress with the new cross-coupling. For now, Molander’s team is advancing its chemistry on its own. “Competition can be good,” but it’s more important to make sure everyone on the playing field is treated fairly, MacMillan says. “This was a win for everyone.”