me-PCR: a refined ultrafast algorithm for identifying sequence-defined genomic elements.

We have adapted the originally described electronic PCR (e-PCR) algorithm to perform string searches more accurately and much more rapidly than previously possible. Our implementation [multithreaded e-PCR (me-PCR)] runs sufficiently fast to allow even desktop machines to query quickly large genomes with very large genomic element sets. In addition, me-PCR is multithreaded, interprets all IUPAC nucleotide symbols, allows searches with elements specified by long sequences (such as SNPs), accepts ranges in the expected PCR size input field, requires substantially less memory for analysis of large sequences and corrects a number of minor flaws causing misreporting of hits in exceptional cases. Thus, me-PCR provides increased annotation capabilities for complex genomes to non-expert laboratories.