H5N1 Sequence Hoarding By The WHO and China
Recombinomics Commentary
November 5, 2006
Virus are always mutating and changing with every new generation and new strains and substrains are always appearing, but what was crucial was to try to understand which strains were becoming dominant, Hall said.
A dominant strain of the bird flu virus could be an indication that it was becoming resistant to vaccines given to poultry. China vaccinated up to 8.2 billion poultry from January to September this year.
"Right now, there is not enough data to make a firm conclusion... but this is what the data is indicating," Hall said.
The WHO would like to see the "hundreds, if not thousands of (bird flu) gene sequences" uploaded by the Ministry of Agriculture on a public data base, she said.
The above comments by the WHO are ironic, since WHO likely has more sequences in its password protected private database than China has withheld. In August, Indonesia asked WHO to release over 700 sequences form human H5N1 isolates, which were password protected on WHO’s private H5N1 database at Los Alamos. These sequences became public instantly when the password protection was removed.
However, the number of sequences that have been withheld by WHO dwarfs these numbers or the numbers mentioned in the quote above regarding sequences in China. The comments above were precipitated by the recently published PNAS paper which has HA and PB2 sequences from 405 isolates collected in China in 2005 and 2006. Since each isolate has eight gene segments, these isolates contained 3240 sequences. However, only 556 sequences were released.
Thus, from this 2005/2006 series alone, 2684 sequences were withheld. Released sequences included 404 full or partial sequences from HA and 152 full sequences from PB2. No sequences were released for six of the eight gene segments, even though amantadine sensitivity of the M2 gene product was discussed in the published paper. Peer review journals generally require desposit of sequences discussed so the data can be independently verified.. Moreover, only a subset of the PB2 sequences were released, and most of the HA sequences were partial sequences.
Similarly, earlier H5N1 sequences from the two labs who generated the above sequences, St Jude and Hong Kong University, have only released partial sequences of H5N1 isolates from Hong Kong, China, and Vietnam, even though these sequences have been published and some date back to 2000.
Remarkably, full sequences can be generated at no cost under the NIAID flu sequencing program, and St Jude has sequenced low path bird and human isolates under that program, but they have not completed any of the published H5N1 sequences under the NIAID program.
The influenza sequencing project, which recently complete full H5N1 sequences on samples submitted by the Capua lab (172 sequences have been released and 40 more are in the validation stage), requires publication of the data in a public database, as requested by WHO.
Most of the samples from the Capua lab were from Qinghai isolates collected in 2006 from countries that sent samples for confirmation by an outside lab. Many countries in Europe sent samples to the WHO affiliated lab in Weybridge. These sequences were placed in the WHO database, but only 8 sequences from one turkey have been released, in addition to 40 human sequences. Data presented at the WHO sponsored meeting in Italy however, discussed over 700 samples from Europe that were H5N1 positive, including the 80 isolates listed on phylogenetic trees. Thus, over 600 European H5N1 sequences have been hoarded and restricted from public access by the WHO database at Los Alamos.
Recently, over 300 sequences from H5N1 in China were released by the Beijing Genome Institute. These releases were from H5N1 isolates from poultry, wild birds, and swine in northern China collected between 1997 and 2004. Clearly, similar full sequences from all eight gene segments from birds from 2005 and 2006 would compliment the sequences that are public, as well as the sequences sequestered in the WHO private database. Moreover, H5N1 sequences from fatal infections in China in 2005 and 2006 have been released, but full sequences are lacking in these isolates as well as additional human and bird isolates described in Chinese Ministry Of Health report from the beginning of this year.
The released sequences contain clear evidence of transmission and tranportaion of H5N1 over a wide geographical region within China and to locations as far away as Europe and Africa. These are also regions of Asian genes in North American isolates, highlighting the role of recombination in the evolution of H5N1 worldwide.
The time for release of the sequences held in China, as well as the WHO private database, is long overdue. WHO should start by releasing the sequences it is hoarding and encourage its consultants to finish the sequences from published isolates and make those data public as soon as possible, which can be done in weeks through the free NIAID influenza sequencing project.
Media Sources
Phylogenetic Trees