Method to Identify Coming Predominant Influenza Strains
Once genetic sequences of influenza viruses become available their evolution can be analyzed. Tracking the evolutionary path is important for the design of seasonal influenza vaccines, but also to investigate the origins and continued evolution of pandemic influenza viruses. Phylogenetic trees should accurately reflect the pathways of evolution, but even more desired would be their ability to accurately identify those newly emerging strains that will eventually rise to predominance in the host population.
Max-Planck Institute for Informatics in Saarbrücken is devoted to cutting-edge research in informatics with a focus on algorithms and their applications. The hereby offered methodology computes a phylogenetic tree and gene-allele predominance plots for fast-evolving populations by integrating not only sequence data, but also by consistently processing time stamp data associated with each isolate in an unprecedented fashion. The resulting phylogenetic tree and gene allele predominance plots allow to accurately identify those distinct viral strains that are likely to be suitable vaccine strain candidates for the following influenza season. This is, because they are found to be prone to rise to predominance then.
Unique methodology allows to identify „in-silico“ the presumably most relevant viral isolates as vaccine strains.
The methodology is not restricted to influenza viruses, but can be applied for the analysis of other fast-evolving genetic entities.
A meanwhile world wide patent application protecting the computer-implemented method is pending since November 25th 2008.
Example: Northern hemisphere allele selection: Predominant alleles were named according to literature names of predominant influenza A (H3N2) variants between 1998 and 2009.The PA99 alleles were first seen in 1998/99 and became predominant in 1999/2000.
- The FU02 were selected in 2002/03 and became predominant in the season 2003/04.
- The first CA04 allels were seen in 2003/04, but the final allels were detected in 2004/05 when they were already predominant.
- WI05 alleles were selected in 2005/06, one year before they became predominant.
- In 2006/07 BR07 alleles were selected that became predominant in 2007/08.
Hence, the selection of the different alleles by the computed method would have worked in 4 out of 5 cases.