12Background: Understanding the genomic basis of phenotypic diversity can be greatly facilitated 13 by examining adaptive radiations with hypervariable traits. In this study, we focus on a rapidly 14 diverged species group of mormyrid electric fish in the genus Paramormyrops, which are 15 characterized by extensive phenotypic variation in electric organ discharges (EODs). The main 16 components of EOD diversity are waveform duration, complexity and polarity. Using an RNA-17 sequencing based approach, we sought to identify gene expression correlates for each of these 18 EOD waveform features by comparing 11 specimens of Paramormyrops that exhibit variation in 19 these features.
20Results: Patterns of gene expression among Paramormyrops are highly correlated, and 3,274 21 genes (16%) were differentially expressed. Using our most restrictive criteria, we detected 71-22 144 differentially expressed genes correlated with each EOD feature, with little overlap between 23 them. The predicted functions of several of these genes are related to extracellular matrix, cation 24 homeostasis, lipid metabolism, and cytoskeletal and sarcomeric proteins. These genes are of 25 significant interest given the known morphological differences between electric organs that 26 underlie differences in the EOD waveform features studied.
27Conclusions: In this study, we identified plausible candidate genes that may contribute to 28 phenotypic differences in EOD waveforms among a rapidly diverged group of mormyrid electric 29 fish. These genes may be important targets of selection in the evolution of species-specific 30 differences in mate-recognition signals.Understanding the genomic basis of phenotypic diversity is a major goal of evolutionary 33 biology [1]. Adaptive radiations and explosive diversification of species [2] are frequently 34 characterized by interspecific phenotypic differences in divergence of few, hypervariable 35 phenotypic traits [3][4][5][6]. Such systems offer exceptional advantages to study the genomic bases of 36 phenotypic diversity: they can provide replication under a controlled phylogenetic framework 37 [7], and couple ample phenotypic differentiation with relatively "clean" genomic signals between 38 recently diverged species [8]. Study of the genomic mechanisms underlying hypervariable 39 phenotypic traits has identified, in some cases relatively simple genetic architectures [9][10][11][12][13]. 40 More often, the genetic architecture underlying such traits can be complex and polygenic [14-41 17]. It has long been recognized that changes in gene expression can affect phenotypic 42 differences between species [18], and RNA-seq based approaches have greatly facilitated the 43 study of this relationship [19]. A growing number of studies have examined differences in gene 44 expression in phenotypic evolution (e.g., [19][20][21][22][23][24][25][26][27]). While these studies do not implicate 45 mutational causes, analysis of differential gene expression (DGE) can be a useful approach in 46 examining the ...