Improvements to pairwise sequence comparison (PASC): a genome-based web tool for virus classification

Abstract

The number of viral genome sequences in the public databases is increasing dramatically, and these sequences are playing an important role in virus classification. Pairwise sequence comparison is a sequence-based virus classification method. A program using this method calculates the pairwise identities of virus sequences within a virus family and displays their distribution, and visual analysis helps to determine demarcations at different taxonomic levels such as strain, species, genus and subfamily. Subsequent comparison of new sequences against existing ones allows viruses from which the new sequences were derived to be classified. Although this method cannot be used as the only criterion for virus classification in some cases, it is a quantitative method and has many advantages over conventional virus classification methods. It has been applied to several virus families, and there is an increasing interest in using this method for other virus families/groups. The Pairwise Sequence Comparison (PASC) classification tool was created at the National Center for Biotechnology Information. The tool's database stores pairwise identities for complete genomes/segments of 56 virus families/groups. Data in the system are updated every day to reflect changes in virus taxonomy and additions of new virus sequences to the public database. The web interface of the tool ( http://www.ncbi.nlm.nih.gov/sutils/pasc/ ) makes it easy to navigate and perform analyses. Multiple new viral genome sequences can be tested simultaneously with this system to suggest the taxonomic position of virus isolates in a specific family. PASC eliminates potential discrepancies in the results caused by different algorithms and/or different data used by researchers.

Fig. 1

Frequency distribution of pairwise identities from the complete genome sequence comparison of 87 microviruses

Fig. 2

Frequency distribution of pairwise identities from the complete genome sequence comparison of 274 caliciviruses, and the simulation of taxonomy changes using proposed species and genus demarcations

Fig. 3

Frequency distribution of pairwise identities from the complete genome sequence comparison of 260 potyviruses and its application in classifying newly sequenced viruses

Fig. 4

Frequency distribution of pairwise identities from the complete genome sequence comparison of 81 baculoviruses, using global alignment (a) and BLAST-based alignment (b and c). The pairwise genome identities are calculated by the average (b) and maximum (c) length

Fig. 5

Dot matrix and text views of pairwise alignment between genome sequences of enterobacteria phage SP6 (NC_004831) and enterobacteria phage T7 (NC_001604), using the BLAST-based and global alignment methods. The conserved regions between the two genomes are marked with arrows on the dot matrix from the BLAST-based alignment