Stoush demonstrates need to discuss data sharing

Fresh from a clash between a marine biologist and a Massachusetts Institute of Technology (MIT) associate about sharing and publishing publicly-available research, a group of scientists is meeting this week at the US National Academy of Sciences to try to reach consensus on the issue.

The latest flap erupted last week when Mitchell Sogin of the Marine Biological Laboratory in Woods Hole, Massachusetts, accused Hyman Hartman of MIT and a Harvard University associate of "borrowing" sequencing data from his public web site. The story first appeared on February 15 in Science magazine.

As a result, Sogin temporarily shut down his web site, which was slated to begin undergoing restoration on Tuesday with, among other things, a new "data release policy." The web site contained sequencing information for Giardia Lamblia, a waterborne pathogen that causes diarrhoea.

"Our data release policy was very specific," Sogin said. "The vast majority [of site visitors] have respected that. As a result of this fiasco, we have made our data release policy explicit."

The dispute arose when Hartman of MIT and Alexei Federov of Harvard co-published the paper "The Origin of the Eukaryotic Cell: A Genomic Investigation" in the February 5 Proceedings of the National Academy of Sciences (PNAS). The paper posits that eukaryotes (cells with nuclei) inherit some cell structures from organisms that no longer exist.

"In my estimation, [Hartman] did a large-scale analysis based on unpublished data without first confirming the accuracy and without consulting us about the accuracy of the data," Sogin charged. "Some people think that because it appears on a web (site), that constitutes some sort of publishing."

Hartman and Federov liberally cite Sogin's research in their paper. Hartman said he even offered to e-mail the paper to Sogin before the corrections phase, but was told to simply drop it in the mail. Then Sogin said he didn't receive it prior to publication in PNAS, according to Hartman.

"From February, last year until January when this paper went on the Internet, he was totally informed of everything I was doing. I did not do anything wrong," said an exasperated Hartman. "I don't want publicity. I want the paper to live. I would never try to use his work unless he knew I was using it. He did not take me seriously. That's his one fault. Other than that, he's a good scientist. He's done great work."

Colleagues, too polite to say so on the record, confided that Sogin is burned because Hartman beat him to punch with his paper. Indeed, when the story broke in Science, the headline said Sogin had been "...Scooped With His Own Data."

PNAS editor-in-chief Nicholas Cozzarelli confirmed Sogin asked him to make certain changes to the Hartman/Federov paper, but he declined, sharply disagreeing with Sogin's position.

"He asked for a lot of things. I refused his request because I felt he's begging for people to use his results and then wants to prevent them from publishing them," Cozzarelli said. "Who gave him that right?"

Cozzarelli has advocated virtually unrestricted use of data published on web sites. Last July, he authored a PNAS editorial headlined "Unfettered Access of Published Results" that criticised a publication for withholding complete access to the draft sequence of the human genome.

At this week's conference, Cozzarelli moderated a case study that asks: "What are an author's obligations to share a reagent on which he or she has published, especially given a rare resource?"