Multiplex primer prediction software for divergent targets

Gardner, Shea N.; Hiddessen, Amy L.; Williams, Peter L.; Hara, Christine; Wagner, Mark C.; Colston, Bill W.

Researchers employ numerous approaches for viral detection and discovery, including metagenomic sequencing (1) (2) (3) , microarrays (4) (5) (6) (7) (8) (9) or multiplex PCR followed by other methods of characterization such as mass spectrometry (10) (11) (12) (13) (14) , suspension arrays (15, 16) or amplicon sequencing (17) . Multiplex PCR followed by analysis of amplified products fills a niche for viral identification when singleplex PCR has failed or there are a few dozen likely candidates but the expense of metagenomic sequencing or high-density microarrays is unwarranted (18) . However, multiplex primer design for many highly divergent targets is challenging since no universally conserved primers may exist, and finding sets of primers likely to function well in multiplex (e.g. isothermal T m 's, no primer dimers) adds to the complexity of finding conserved primer candidates. Primer design software that requires a multiple sequence alignment (MSA) as input can be problematic for diverse target sets, as MSAs can be difficult to construct, exhausting memory or available time before an alignment is completed. Even if an alignment does complete for divergent target sets such as all members of a family of RNA viruses or gene homologues across species, alignments may show little nucleotide sequence conservation, and multiple primers are required to amplify all targets. Considering the challenges of primer design for targets showing sequence variation, it is not surprising that many of the PCR-based assays in the literature are predicted to fail to detect desired targets when compared against available sequence data, and this problem is worst at higher taxonomic levels like family (19) .