Multiplex primer prediction software for divergent targets

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

The increase in sequence data requires $700 more 10-mer primers to amplify all sequenced viruses in 2007 compared to 2004 (Figure 2 ). While the increase in the number of sequences used between the two dates was only $15%, the number of primers required increased by 48%, illustrating the substantial increase in diversity represented by the additional sequence data. Figure 2 also shows that removing all T m constraints allows fewer primers to be used since no conserved primers are eliminated due to T m , as some AT rich subsequences tend to be fairly conserved. Figure 3 shows that a universal primer set predicted using the 2004 sequence data would amplify only 35% of the 2007 sequences using primers of at least 10 nt in length. Shorter primers increase this fraction to over 60%, due to the higher likelihood of occurrence and conservation of shorter oligos, but even so, a multiplex of 7-mers is not guaranteed to amplify a fragment from every virus. The minor differences in the number of genomes detected between 12, 13, 14 and 15-mers can be attributed to the facts that a greedy but not necessarily optimal algorithm is used to select one solution from among many, that the primers in a particular set depend on the T m ranges we used as well as length differences, and the unpredictable nature of novel viral sequences accumulated between 2004 and 2007, rather than any real difference in the ability to detect genomes among those primer sets.