Chinese medical workers take part in a drill to simulate an outbreak of H7N9 flu in the population. Credit: ChinaFotoPress/Photoshot

On 31 March, China reported the first human cases of infection with a new H7N9 avian flu virus. The same day, a team at the Chinese National Influenza Center (CNIC) in Beijing uploaded to a research database the genetic sequences of viruses isolated from the first three human cases. But Nature has learned that in the days that followed, Chinese scientists and officials grew increasingly concerned that China might lose credit for its work in isolating and sequencing the virus.

The sequences were placed in the Global Initiative on Sharing All Influenza Data (GISAID) database. According to the database’s rules, scientists who use sequences from it must credit those who deposited the data and, where possible, propose collaborations with them.

“Unfortunately some bad things happened when we released the sequences in GISAID, and they really hurt us,” says Yuelong Shu, head of the CNIC, which is also the World Health Organization (WHO) Collaborating Centre for Reference and Research on Influenza in China. “GISAID have tried their best to help us,” he adds. “I really appreciate what they have done.”

Shu did not initially reveal specific concerns, but other researchers have told Nature some of the details. On 5 April, the Chinese scientists submitted their first major H7N9 paper, including analyses of the sequences, to The New England Journal of Medicine (NEJM).

Around the same time, the researchers learned that they might be scooped: several other research groups were preparing to publish papers on the virus, or already had done so, including analyses of the sequences in GISAID.

This news was followed by what seemed to be a snub. It emerged on 5 April that drug firm Novartis in Basel, Switzerland, and the J. Craig Venter Institute in Rockville, Maryland, planned to use the uploaded sequences to develop H7N9 vaccines. The initiative had US government funding and would be a collaboration with the US Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia — but not with the Chinese team. The Chinese researchers felt that this was not in the spirit of GISAID.

The sharing of flu-virus data and materials has long been a politically charged issue in global health. Timely information from potentially pandemic flu strains is crucial for efforts to monitor drug resistance and the evolution of viruses, and for the development of diagnostics and vaccines. But some countries have been reluctant to share such data because they have seen little in return in terms of collaboration, technology transfer or access to the drugs and vaccines developed as a result.

GISAID was created in 2008 to help overcome some of these concerns. “Without a mechanism like GISAID it would be very difficult for various authorities to make information available prior to publication,” says Alan Hay, co-chair of GISAID’s scientific advisory council.

Novartis spokeswoman Liz Power says that after the company’s researchers downloaded the H7N9 sequences, it “explored research collaboration” with the Chinese CDC in Beijing, of which the CNIC is part. “We are committed to sharing any meaningful insights coming out of our work with China CDC,” she says.

Kristine Sheedy, a spokeswoman for the US CDC, acknowledges that “there were differences in understanding and expectations regarding use of the Chinese H7N9 sequence data by several outside groups”, but adds that the US CDC was not among them. The CDC has had a strong ongoing collaboration with its counterpart in China since the start of the H7N9 outbreaks, she says.

Shu says that the Chinese researchers would have preferred for the vaccine developers to have told them in advance about how they intended to use the sequences, but adds that communication channels have now been opened and that the various parties have agreed to collaborate. “Thanks to the president of GISAID this situation was quickly mitigated,” says Shu.

Chinese worries over being scooped also seem to have been put to rest. The CNIC scientists were most concerned about a major analysis of the H7N9 virus scheduled for publication on 10 April in Eurosurveillance — which would have appeared before the Chinese NEJM paper.

A co-author on the Eurosurveillance paper, virologist Masato Tashiro of the Influenza Virus Research Center in Tokyo, the WHO’s influenza reference centre in Japan, says that he sent a draft of the paper to the Chinese researchers on 7 April, inviting them to be co-authors. They declined, but asked Tashiro to delay publication until after their NEJM paper had appeared. He agreed and the NEJM paper was published on 11 April (R. Gaoet al.N.Engl.J.Med.http://doi.org/k7r;2013), with the Euro­surveillance paper appearing later the same day (T. Kage­yama et al. Euro Surveill. 18, 20453; 2013). Tashiro notes that holding the paper did not have an impact on public health, because all its analyses were shared with the WHO’s global network of flu labs on 1 and 2 April, and were used to help the WHO to prepare its initial risk assessment of the virus (see ‘Flu tracking’).

boxed-text

“One has to recognize the sensitivities in relation to scientific priority,” says Hay, who thinks that many potential difficulties could be avoided if people spoke to each other more about their work and their publication plans.

“Scientific etiquette is without doubt a key to keeping the rapid sharing of data a reality,” says Shu. In this case, he continues, “after some initial concerns we found that both researchers and publishers were understanding of our predicament”.

Hay hopes that the hiccups won’t discourage Chinese researchers from making their H7N9 data publicly accessible quickly. “It is very important to continue to share sequences from the more recent cases,” he says.

For his part, Shu says that he is keen to ensure that researchers continue to have “unfettered access to data”.