Defining objective clusters for rabies virus sequences using affinity propagation clustering. Defining objective clusters for rabies virus sequences using affinity propagation clustering

Rabies is caused by lyssaviruses, one of the oldest known zoonoses. In recent years, more than 21,000 nucleotide sequences of rabies viruses (RABV), from the prototype species rabies lyssavirus (RABLV), have been deposited in public databases. Subsequent phylogenetic analyses in combination with metadata suggest space-dependent distributions of RABV. However, these analyses were limited by the lack of objective criteria for cluster allocations in phylogenetic trees. Therefore, a mathematical approach named ‘affinity propagation clustering’ (AP) was applied in order to eliminate this limitation and to propose a way to classify full-genome RABV sequences. Because of its computational properties, AP has already been successfully applied for other tasks in bioinformatics, such as analysis of microarray and gene expression data, but not very broadly for cluster analysis in sequence comparisons. Existing (516) and original (46) full genome RABV sequences were used to demonstrate the application of AP for RABV clustering. On a global scale, AP proposed four clusters, i.e. New World cluster, Arctic/Arctic-like, Cosmopolitan, and Asian as previously assigned by phylogenetic studies. In summary, we combine AP with established phylogenetic analyses, which is possible to resolve phylogenetic relationships between objectively determined clusters and sequences. This workflow may be useful in confirming cluster distributions in a transparent manner, not only for RABV, but also for other comparative sequence analyses.