Abstract
Fishes in the cypriniform family Catostomidae (suckers) are evolutionary tetraploids. The use of nuclear markers in the phylogenetic study of this important group has been greatly hindered by the challenge of identifying paralogous copies of genes. In the present study, we used two different methods to separate the gene copies of five single-copy nuclear genes (i.e., RAG1, EGR2B, EGR3, IRBP2, and RAG2). For each gene, all sequences of Copy I formed a clade that was sister to the clade formed by all sequences of Copy II in the phylogenetic trees. The maternal and paternal progenitor of the tetraploid ancestor of the Catostomidae could not be determined. We also constructed a mitochondrial tree to reflect the maternal relationships among major catostomid lineages. Our data appear to support a sister relationship between Catostominae and a monophyletic group composed of Myxocyprininae, Cycleptinae, and Ictiobinae. However, within Catostominae, there is significant conflict between mitochondrial and nuclear data regarding the relationships among Erimyzonini, Catostomini, and Moxostomatini/Thoburnini. Many indels, unexpected stop codons, and possible gene loss were identified in one gene copy of RAG1, RAG2, and IRBP2. We believe that additional nuclear genome data are needed to better resolve the phylogenetic relationships within the family Catostomidae.