Date of this Version
Background: Exon-primed intron-crossing (EPIC) markers have three advantages over anonymous genomic sequences in studying evolution of natural populations. First, the universal primers designed in exon regions can be applied across a broad taxonomic range. Second, the homology of EPIC-amplified sequences can be easily determined by comparing either their exon or intron portion depending on the genetic distance between the taxa. Third, having both the exon and intron fragments could help in examining genetic variation at the intraspecific and interspecific level simultaneously, particularly helpful when studying species complex. However, the paucity of EPIC markers has hindered multilocus studies using nuclear gene sequences, particularly in teleost fishes.
Results: We introduce a bioinformatics pipeline for developing EPIC markers by comparing the whole genome sequences between two or more species. By applying this approach on five teleost fishes whose genomes were available in the Ensembl database http://www.ensembl.org, we identified 210 EPIC markers that have single-copy and conserved exon regions with identity greater than 85% among the five teleost fishes. We tested 12 randomly chosen EPIC markers in nine teleost species having a wide phylogenetic range. The success rate of amplifying and sequencing those markers varied from 44% to 100% in different species. We analyzed the exon sequences of the 12 EPIC markers from 13 teleosts. The resulting phylogeny contains many traditionally well-supported clades, indicating the usefulness of the exon portion of EPIC markers in reconstructing species phylogeny, in addition to the value of the intron portion of EPIC markers in interrogating the population history.
Conclusions: This study illustrated an effective approach to develop EPIC markers in a taxonomic group, where two or more genome sequences are available. The markers identified could be amplified across a broad taxonomic range of teleost fishes. The phylogenetic utility of individual markers varied according to intron size and amplifiability. The bioinformatics pipelines developed are readily adapted to other taxonomic groups.