The genome of the Gulf pipefish enables understanding of evolutionary innovations

Abstract

BackgroundEvolutionary origins of derived morphologies ultimately stem from changes in protein structure, gene regulation, and gene content. A well-assembled, annotated reference genome is a central resource for pursuing these molecular phenomena underlying phenotypic evolution. We explored the genome of the Gulf pipefish (Syngnathus scovelli), which belongs to family Syngnathidae (pipefishes, seahorses, and seadragons). These fishes have dramatically derived bodies and a remarkable novelty among vertebrates, the male brood pouch.ResultsWe produce a reference genome, condensed into chromosomes, for the Gulf pipefish. Gene losses and other changes have occurred in pipefish hox and dlx clusters and in the tbx and pitx gene families, candidate mechanisms for the evolution of syngnathid traits, including an elongated axis and the loss of ribs, pelvic fins, and teeth. We measure gene expression changes in pregnant versus non-pregnant brood pouch tissue and characterize the genomic organization of duplicated metalloprotease genes (patristacins) recruited into the function of this novel structure. Phylogenetic inference using ultraconserved sequences provides an alternative hypothesis for the relationship between orders Syngnathiformes and Scombriformes. Comparisons of chromosome structure among percomorphs show that chromosome number in a pipefish ancestor became reduced via chromosomal fusions.ConclusionsThe collected findings from this first syngnathid reference genome open a window into the genomic underpinnings of highly derived morphologies, demonstrating that de novo production of high quality and useful reference genomes is within reach of even small research groups.Electronic supplementary materialThe online version of this article (doi:10.1186/s13059-016-1126-6) contains supplementary material, which is available to authorized users.

Genome‐wide selection components analysis in a fish with male pregnancy

“…Raw reads were filtered for quality and grouped by barcode using the Stacks version 1.29 function process_radtags (Catchen et al 2011;Catchen et al 2013). Bowtie 2.0 (Langmead and Salzberg 2012) was used to map the filtered reads to a draft S. scovelli genome assembly (Small et al 2016), using the default "sensitive" parameters (--sensitive). A catalog of RAD loci was created and individuals were genotyped using the ref_map.pl program in Stacks version 1.37 (Catchen et al 2011;Catchen et al 2013).…”

Comparative transcriptomics of cyprinid minnows and carp in a common wild setting: a resource for ecological genomics in freshwater communities

“…Our interest in these genes lies in the idea that they may contain some of the functional elements responsible for extensive ecological and phylogenetic diversity present in the Cyprinidae, as in previous studies of lineage-specific gene family expansions 83–85 . For example, expansion of the patristacin gene family in pipefish may be an important driver in the evolution of male pregnancy in that lineage 85 . Many of the potentially novel genes we identified may prove to be false positives as more fish genomes are sequenced and annotated; however, these candidates would be an excellent starting point for researchers interested in targeted searches for genes or proteins underlying ecological novelty in cyprinids that may have arisen through local gene duplications, exon shuffling, horizontal transfer, or other mechanisms.…”

Substantial differences in bias between single‐digest and double‐digest RAD‐seq libraries: A case study

“…The raw reads from each sequencing run were separated by barcode using the process_radtags module of Stacks (Catchen, Amores, Hohenlohe, Cresko, & Postlethwait, ; Catchen, Hohenlohe, Bassham, Amores, & Cresko, ), and each individual's reads were aligned to the Gulf pipefish genome (Small et al., ) using Bowtie 2.0 (Langmead & Salzberg, ) with the –sensitive parameters. The genome contains 22 major linkage groups, corresponding to the 22 chromosomes of S. scovelli .…”

“…We used gbstools (Cooke et al., ) to estimate SNPs with high rates of restriction site allelic dropout. After adding MboI cut sites to the gbstools scripts, we used gbstools to digest the S. scovelli genome (Small et al., ) using both Pst I and Mbo I (to estimate allelic dropout in the ddRAD data set) and using only Pst I (to estimate allelic dropout in the sdRAD data set). We then used GBStools and python v. 2.7 to estimate allelic dropout in the combined RAD‐seq data set using normalization factors of 1.0.…”