Dmitry Dernovoy

Influenza epidemics cause morbidity and mortality worldwide (4). Each year in the United States, more than 200,000 patients are admitted to hospitals because of influenza and there are approximately 36,000 influenza-related deaths (14). In recent years, several subtypes of avian influenza viruses have jumped host species to infect humans. The H5N1 subtype, in particular, has been reported in 328 human cases and has caused 200 human deaths in 12 countries (World Health Organization, http://www.who.int/csr/disease/avian_influenza/country/cases_table_2007_09_10/en/index.html). These viruses have the potential to cause a pandemic in humans. Antiviral drugs and vaccines must be developed to minimize the damage that such a pandemic would bring. To achieve this, it is vital that researchers have free access to viral sequences in a timely fashion, and sequence analysis tools need to be readily available. Historically, the number of influenza virus sequences in public databases has been far less than those of some well-studied viruses, such as human immunodeficiency virus. The number of complete influenza virus genomes has been even smaller. In addition, many of the sequences were collected in the course of influenza surveillance programs that prioritized antigenically novel isolates. Although collecting antigenically novel isolates is appropriate for surveillance, it results in biased samples of sequenced isolates that are not representative of community cases of influenza (2, 13). Therefore, in 2004, the National Institute of Allergy and Infectious Diseases (NIAID) launched the Influenza Genome Sequencing Project (7), which aims to rapidly sequence influenza viruses from samples collected all over the world. Viral sequences were generated at the J. Craig Venter Institute, annotated at the National Center for Biotechnology Information (NCBI), and deposited in GenBank. In just over 2 years after the initiation of the project, more than 2,000 complete genomes of influenza viruses A and B had been deposited in GenBank. To help the research community to make full use of the wealth of information from such a large amount of data, which will be increasing continuously, the Influenza Virus Resource was created at NCBI in 2004.

Influenza viruses are remarkably adept at surviving in the human population over a long timescale. The human influenza A virus continues to thrive even among populations with widespread access to vaccines, and continues to be a major cause of morbidity and mortality. The virus mutates from year to year, making the existing vaccines ineffective on a regular basis, and requiring that new strains be chosen for a new vaccine. Less-frequent major changes, known as antigenic shift, create new strains against which the human population has little protective immunity, thereby causing worldwide pandemics. The most recent pandemics include the 1918 'Spanish' flu, one of the most deadly outbreaks in recorded history, which killed 30-50 million people worldwide, the 1957 'Asian' flu, and the 1968 'Hong Kong' flu. Motivated by the need for a better understanding of influenza evolution, we have developed flexible protocols that make it possible to apply large-scale sequencing techniques to the highly variable influenza genome. Here we report the results of sequencing 209 complete genomes of the human influenza A virus, encompassing a total of 2,821,103 nucleotides. In addition to increasing markedly the number of publicly available, complete influenza virus genomes, we have discovered several anomalies in these first 209 genomes that demonstrate the dynamic nature of influenza transmission and evolution. This new, large-scale sequencing effort promises to provide a more comprehensive picture of the evolution of influenza viruses and of their pattern of transmission through human and animal populations. All data from this project are being deposited, without delay, in public archives.

The distributed-genome hypothesis (DGH) states that pathogenic bacteria possess a supragenome that is much larger than the genome of any single bacterium and that these pathogens utilize genetic recombination and a large, noncore set of genes as a means of diversity generation. We sequenced the genomes of eight nasopharyngeal strains of Streptococcus pneumoniae isolated from pediatric patients with upper respiratory symptoms and performed quantitative genomic analyses among these and nine publicly available pneumococcal strains. Coding sequences from all strains were grouped into 3,170 orthologous gene clusters, of which 1,454 (46%) were conserved among all 17 strains. The majority of the gene clusters, 1,716 (54%), were not found in all strains. Genic differences per strain pair ranged from 35 to 629 orthologous clusters, with each strain's genome containing between 21 and 32% noncore genes. The distribution of the orthologous clusters per genome for the 17 strains was entered into the finite-supragenome model, which predicted that (i) the S. pneumoniae supragenome contains more than 5,000 orthologous clusters and (ii) 99% of the orthologous clusters ( approximately 3,000) that are represented in the S. pneumoniae population at frequencies of >or=0.1 can be identified if 33 representative genomes are sequenced. These extensive genic diversity data support the DGH and provide a basis for understanding the great differences in clinical phenotype associated with various pneumococcal strains. When these findings are taken together with previous studies that demonstrated the presence of a supragenome for Streptococcus agalactiae and Haemophilus influenzae, it appears that the possession of a distributed genome is a common host interaction strategy.

FLAN (short for FLu ANnotation), the NCBI web server for genome annotation of influenza virus (http://www.ncbi.nlm.nih.gov/genomes/FLU/Database/annotation.cgi) is a tool for user-provided influenza A virus or influenza B virus sequences. It can validate and predict protein sequences encoded by an input flu sequence. The input sequence is BLASTed against a database containing influenza sequences to determine the virus type (A or B), segment (1 through 8) and subtype for the hemagglutinin and neuraminidase segments of influenza A virus. For each segment/subtype of the viruses, a set of sample protein sequences is maintained. The input sequence is then aligned against the corresponding protein set with a 'Protein to nucleotide alignment tool' (ProSplign). The translated product from the best alignment to the sample protein sequence is used as the predicted protein encoded by the input sequence. The output can be a feature table that can be used for sequence submission to GenBank (by Sequin or tbl2asn), a GenBank flat file, or the predicted protein sequences in FASTA format. A message showing the length of the input sequence, the predicted virus type, segment and subtype for the hemagglutinin and neuraminidase segments of Influenza A virus will also be displayed.

As more and more complete bacterial genome sequences become available, the genome annotation of previously sequenced genomes may become quickly outdated. This is primarily due to the discovery and functional characterization of new genes. We have reannotated the recently published genome of Shewanella oneidensis with the following results: 51 new genes have been identified, and functional annotation has been added to the 97 genes, including 15 new and 82 existing ones with previously unassigned function. The identification of new genes was achieved by predicting the protein coding regions using the HMM-based program GeneMark.hmm. Subsequent comparison of the predicted gene products to the non-redundant protein database using BLAST and the COG (Clusters of Orthologous Groups) database using COGNITOR provided for the functional annotation.