Multiplex Degenerate Primer Design for Targeted Whole Genome Amplification of Many Viral Genomes

Gardner, Shea N.; Jaing, Crystal J.; Elsheikh, Maher M.; Peña, José; Hysom, David A.; Borucki, Monica K.

JCVI has an automated degenerate PCR primer design system called JCVI Primer Designer, which is similar to run tiled primers in that it designs degenerate primers to tile across viral genomes [4] . The major difference is that it begins with a consensus sequence containing degenerate bases and selects primers with fewer than 3 or 4 degenerate bases, so that in the end a majority of strains are amplified, 2 Advances in Bioinformatics Table 1 : Summary of average lengths, number of sequences, and percentage of conserved bases in a multiple sequence alignment (with MUSCLE [5] ), and number of tiled primers required for the short and long amplicon settings. but it does not require primers to amplify all strains. In their examples, most of the primer pairs could amplify >75% of isolates. Each primer pair for a given region is intended to be run as a specific pair, not as a multiplex with multiple pairs. Consensus sequences with too little conservation, that is, <90% consensus, are divided manually in a preprocessing step into subgroups which can be run separately through the pipeline. The method here differs in that it takes the full multiple sequence alignment as input rather than a consensus, and it seeks to automatically design a minimal, degenerate set of multiplex compatible primers to amplify all the strains for a given region in a single reaction. The major operational difference of run tiled primers compared to the JCVI pipeline is that run tiled primers does not require manual subdivision of the target sequences into high consensus groups to be run separately by the user, and run tiled primers attempts to cover 100% of the target sequences in a single pass using a greedy minimal set algorithm. Some regions of high conservation may have only one primer pair predicted to amplify all strain variants, while other regions may require many primers to cover all known variants. If multiple strains are present at once or if multiple forward and/or reverse primers in the multiplex amplify the strain present, the reaction will generate multiple overlapping amplicons spanning the same region, which could be problematic if exactly one amplicon sequence is needed, for example, for Sanger sequencing. In this case, the JCVI Primer Designer would be preferable since it designs primer pairs each to be run in singleplex reactions rather than as a multiplex, with the risk that outlier strains may not be amplified. However, when multiple overlapping reads with different endpoints or from different strains are acceptable, as in high throughput sequencing, run tiled primers should be suitable and could serve as a good alternative to random amplification when more specific enrichment is needed, and amplification of outliers is desired.