Abstract

The mangrove rivulus ( Kryptolebias marmoratus ) is one of two preferentially self-fertilizing hermaphroditic vertebrates. This mode of reproduction makes mangrove rivulus an important model for evolutionary and biomedical studies because long periods of self-fertilization result in naturally homozygous genotypes that can produce isogenic lineages without significant limitations associated with inbreeding depression. Over 400 isogenic lineages currently held in laboratories across the globe show considerable among-lineage variation in physiology, behavior, and life history traits that is maintained under common garden conditions. Temperature mediates the development of primary males and also sex change between hermaphrodites and secondary males, which makes the system ideal for the study of sex determination and sexual plasticity. Mangrove rivulus also exhibit remarkable adaptations to living in extreme environments, and the system has great promise to shed light on the evolution of terrestrial locomotion, aerial respiration, and broad tolerances to hypoxia, salinity, temperature, and environmental pollutants. Genome assembly of the mangrove rivulus allows the study of genes and gene families associated with the traits described above. Here we present a de novo assembled reference genome for the mangrove rivulus, with an approximately 900 Mb genome, including 27,328 annotated, predicted, protein-coding genes. Moreover, we are able to place more than 50% of the assembled genome onto a recently published linkage map. The genome provides an important addition to the linkage map and transcriptomic tools recently developed for this species that together provide critical resources for epigenetic, transcriptomic, and proteomic analyses. Moreover, the genome will serve as the foundation for addressing key questions in behavior, physiology, toxicology, and evolutionary biology.

Introduction

The mangrove rivulus fish, Kryptolebias marmoratus is renowned for being one of two self-fertilizing hermaphroditic vertebrates; the other is its putative sister species, Kryptolebias hermaphroditus (formerly Kryptolebias ocellatus ; [ Costa 2011 ]). These species are distributed in close association with red mangrove forests ( Rhizophora mangle ) along the coastal regions of central and south Florida, the Caribbean and the eastern coasts of Central and South America ( Costa 2006 ). Mangrove rivulus exist primarily as hermaphrodites with preferential internal self-fertilization ( Harrington 1961 ), but the species exhibits environmental sex determination, with primary males developing from embryos incubated at low temperatures (18–20 °C; [ Harrington 1967 ; Harrington and Kallman 1968 ; Ellison et al. 2015 ]), and remains sexually labile throughout adulthood ( Harrington 1971 , 1975 ; Turner et al. 2006 ; Garcia et al. 2016 ) ( fig. 1 ). No functional females have been identified in the wild or in captivity ( Harrington 1961 , 1975 ; Farmer and Orlando 2012 ) leading to their classification as an androdioecious vertebrate; however, histological studies have shown that some animals possess only ovarian tissue when young ( Cole and Noakes 1997 ), leading to their classification as an androdioecious vertebrate. Male abundance varies among populations from 0% to 24% ( Turner, Davis, et al. 1992 ) and there is evidence for correspondingly variable outcrossing rates in some locales (e.g., Central America; Tatarenkov et al. 2015 ). While males have not been observed in all populations, microsatellite data from wild-caught individuals successfully identified highly homozygous lineages as well as individuals with heterozygosity expected under random mating, suggesting that outcrossing between hermaphrodites and males is common in some populations ( Taylor et al. 2001 ; Mackiewicz, Tatarenkov, Taylor, et al. 2006 ; Earley et al. 2012 ); hermaphrodites do not mate with one another ( Furness et al. 2015 ).

F ig . 1.—

Photograph of hermaphrodite (top) and secondary male (bottom). Secondary males lack ocellus and have a pinker coloration than hermaphrodites. Photo credits to D. Scott Taylor.

F ig . 1.—

Photograph of hermaphrodite (top) and secondary male (bottom). Secondary males lack ocellus and have a pinker coloration than hermaphrodites. Photo credits to D. Scott Taylor.

These features of mangrove rivulus biology have motivated significant efforts into understanding the environmental ( Harrington 1967 , 1968 ), and physiological/molecular (e.g., [ Kanamori et al. 2006 ; Orlando et al. 2006 ; Park et al. 2013 ]) mechanisms underlying sex determination. The coupling of these features with multiple genetically distinct lineages for many populations, evidence for strong population genetic structure ( Tatarenkov et al. 2007 , 2015 ), and variation among lineages and populations in the extent to which environmental factors induce male development ( Harrington 1968 , 1975 ; Kristensen 1970 ; Ellison et al. 2015 ) position mangrove rivulus as a strong model for understanding sex ratio evolution and the evolutionary ecology of sexual plasticity (e.g., [ Earley et al. 2012 ; Ellison et al. 2015 ]).

Mangrove rivulus have been maintained in laboratory settings for over 50 years ( Harrington 1961 ). There are currently over 400 distinct lineages being maintained in laboratories throughout the globe (Earley R, personal communication), and genetic divergence among lineages has been assessed primarily via microsatellite typing and restriction-site associated DNA sequencing approaches ( Mackiewicz, Tatarenkov, Turner et al. 2006 ; Tatarenkov et al. 2010,, 2012 ). These lineages are easily reared, reproduce readily via self-fertilization when individually housed, and have a relatively short generation time (ca. 3–5 months). Wild-caught lineages that are highly heterozygous can, if given the opportunity only to self, produce offspring that are homozygous at all microsatellite loci after approximately 5–10 generations ( Mackiewicz, Tatarenkov, Perry, et al. 2006 ), which effectively generates many isogenic lineages with different allele combinations of the progenitor. Completely homozygous genotypes occur naturally as well ( Kallman and Harrington 1964 ; Taylor et al. 2001 ; Mackiewicz, Tatarenkov, Taylor, et al. 2006 ; Mackiewicz, Tatarenkov, Turner, et al. 2006 ; Tatarenkov et al. 2007 , 2012, 2015 ), which allows for the immediate production of isogenic lineages if those animals exclusively self-fertilize in the laboratory. The reproductive biology of mangrove rivulus provides a unique opportunity to study, with extraordinary resolution, the contribution of genes, environment, and gene-by-environment interactions to the developing phenotype and simplifies quantitative genetic approaches to investigating phenotypic evolution. Genetically identical animals can be maintained in different environments to examine phenotypically plastic responses to an arsenal of ecologically relevant contexts (e.g., salinity, temperature: Lin and Dunson 1999 ; air exposure: Wells et al. 2015 ) and to construct lineage-specific reaction norms ( Earley et al. 2012 ). Furthermore, high levels of genetic diversity within and among populations ( Turner, Elder, et al. 1992 ; Laughlin et al. 1995 ; Lubinski et al. 1995 ; Taylor 2001 ; Mackiewicz, Tatarenkov, Taylor, et al. 2006 ; Mackiewicz, Tatarenkov, Turner, et al. 2006 ; Tatarenkov et al. 2007 , 2012, 2015 ) allow for the study of heritable phenotypic variation among lineages, including the potential for heritable variation in reaction norm characteristics ( Earley et al. 2012 ).

Indeed, virtually all studies that have employed different lineages demonstrate, under common garden conditions, significant among-lineage variation for phenotypic characteristics. These range from sensitivity of the sexual phenotype to early-life or adult temperature regimes ( Harrington 1967 , 1968, 1971 , 1975 ; Kristensen 1970 ; Turner et al. 2006 ; Ellison et al. 2015 ), some aspects of endocrine function (e.g., [ Earley and Hsu 2008 ; Earley et al. 2013 ]), risk-taking and exploratory behavior ( Edenbrow and Croft 2011 , 2012 ), performance in aggressive contests ( Hsu et al. 2008 ), the propensity to voluntarily jettison from the water ( Turko et al. 2011 ), and life history traits ( Garcia et al. 2015 ). These data indicate that there is considerable heritable variation among genotypes that persists for generations under laboratory conditions.

Another key aspect of mangrove rivulus biology is that they exhibit a number of remarkable adaptations to living in the notoriously variable, sometimes noxious, upland mangrove habitat. These environments are tidal, which can leave the fish stranded on moist land or in stagnant crab burrows for significant periods of time, and can expose the fish to salinities ranging from 0 to 60 parts per thousand (ppt), low dissolved oxygen levels, high hydrogen sulfide levels, and extreme temperatures ( Taylor 2012 ). Mangrove rivulus can live for extended periods of time on land (up to 2 months), often occupying rotting logs and leaf litter ( Taylor et al. 2008 ), and will jettison from water in response to changes in their aquatic environment, which they perceive via a number of specialized mechanisms (e.g., neuroepithelial cells; Robertson et al. 2015 ). These fish can navigate on land using a variety of intricate movements ( Pronko et al. 2013 ), some that involve the fish effectively turning itself into a projectile to launch off of the substrate ( Ashley-Ross et al. 2014 ). Emersion onto land also can function in adult thermoregulation ( Gibson et al. 2015 ), and mangrove rivulus will deposit fertilized eggs on land, which appears to reduce oxygen consumption without detrimental consequences to the adult phenotype ( Wells et al. 2015 ). When on land, mangrove rivulus deploy a number of physiological changes that promote aerial respiration, osmoregulation, and ionoregulation ( Wright 2012 ; Turko et al. 2014 ) and they remodel their gills in ways that reduce surface area for gas exchange ( Wright 2012 ).

The coastal habitats of mangrove rivulus, which are being developed at an extraordinary rate, are subject to significant anthropogenic disturbance, particularly with respect to the influx of environmental pollutants (e.g., endocrine disruptors) from urban run-off and wastewater treatment plants. Research into the effects of such pollutants has demonstrated staggering, often sex-specific, effects on gene expression, endocrine function, reproduction, and survival ( Lee et al. 2006,, 2007 ; Rhee et al. 2008 , 2010, 2011 ; Rhee, Lee, et al. 2009 ; Rhee, Kang, et al. 2009 ; Farmer and Orlando 2012 ; Johnson et al. 2016 ), leading to rivulus being promoted as an important model system for ecotoxicology ( Davis 1986 ; Lee et al. 2008 ). Most studies on rivulus’ physiological and locomotory adaptations to extreme environments and on their responses to environmental pollutants use one or very few lineages, which limits our ability to understand the full scope of variation on which selection can act to drive genetic and phenotypic divergence among populations. Another limitation is access to the genetic and genomic resources necessary to provide a framework for our understanding of both the mechanisms underlying and the evolution of the whole-organism phenotype.

A linkage map is now available ( Kanamori et al. 2016 ) and N -ethyl N -nitrosourea (ENU) mutagenesis screens have been successful in this species ( Moore et al. 2012 ; Sucar et al. 2016 ). A transcriptome was recently made available, and shed light on genes that are up- and downregulated during embryonic diapause (the embryo is fully developed but hatching is delayed; [ Mesak et al. 2015 ]). In addition, there has been some success with producing crosses between phenotypically distinct rivulus lineages ( Harrington 1971 ; Mackiewicz, Tatarenkov, Perry, et al. 2006 ; Nakamura et al. 2008 ). One cross between lineages with divergent growth patterns showed that F2s have intermediate phenotypes ( Nakamura et al. 2008 ). Although outcrossing frequencies are low, even under controlled laboratory conditions (e.g., 6–8% Mackiewicz, Tatarenkov, Perry, et al. 2006 ; in vitro: Nakamura et al. 2008 ), there is great promise for generating crosses between phenotypically dissimilar lineages that could produce an arsenal of recombinant inbred lines that would rival the resources available for Arabidopsis thaliana, Caenorhabditis elegans , and Drosophila melanogaster ( Abzhanov et al. 2008 ; Mackay et al. 2009 , 2014 ; Lehner 2013 ; Rankin 2015 ). Also, our growing understanding of embryogenesis ( Mourabit et al. 2011 ; Mourabit and Kudoh 2012 ) will facilitate genetic manipulation during early development.

Here, we present the annotated genome sequence for the mangrove rivulus from a lab-reared population of the Reckley Hill Lake (RHL) lineage from San Salvador, Bahamas. We annotated 27,328 protein-coding genes. We compared the K. marmoratus reference genome to genomic data generated for the sister species K. hermaphroditus (see fig. 2 for the relationship of Kryptolebias to several model fish species) . The availability of a genome assembly for K. marmoratus provides the basis for future comparative genomic and phenotypic studies and, given increased availability of wild-caught isogenic lineages and information about population genetic structure, for highly controlled studies on environmental sex determination, the evolution of phenotypic plasticity, and adaptations to extreme environments.

F ig . 2.—

Cladogram to represent the relationship among K. marmoratus, K. hermaphroditus and several well-known fish species. The origin of hermaphroditism is noted on the cladogram.

F ig . 2.—

Cladogram to represent the relationship among K. marmoratus, K. hermaphroditus and several well-known fish species. The origin of hermaphroditism is noted on the cladogram.

Results

Genome Sequencing, Assembly, and Annotation

The K. marmoratus genome is composed of 24 haploid chromosomes ( Scheel 1972 ), and the genome size was estimated to be 0.936 picograms, based on flow cytometry (Kiehart R, Goddard K, and Turner B, unpublished data) which translates to approximately 915 megabase pairs (Mb). The genome size estimate was consistent with the k-mer based estimate from the raw genomic reads, which estimates the genome size to be 830 Mb. A total of 289,349,170 paired-end (2 × 101 bp) reads were included in the short read assembly. Additionally, mate-pair reads were trimmed and filtered for the adapter using Ion Torrent PGM software for a total of 15,457,550 mate-pair reads postfiltering with an average read length of 87 bp ( supplementary table S1 , Supplementary Material online). The reads were assembled and the resulting draft genome consists of 40,758 scaffolds, for a total of 654 Mb assembled with 3.77% ambiguous bases ( table 1 ). The assembled genome size is similar to the genome size estimated from the sum of contigs from two other de novo genome assemblies (642 Mb [ Mesak et al. 2015 ] and 633 Mb [ Rhee and Lee 2014 ]). The longest scaffold is 1,759,740 bp and the N50 scaffold length is 111,539 bp. The de novo genome assembly was combined with the recently published linkage map ( Kanamori et al. 2016 ). Of the 9,904 markers used to construct the linkage map, 9,855 markers map uniquely (99.2% of markers) to 4,871 scaffolds, 21 do not map to the genome assembly, and 28 do not map uniquely (map to two positions). More than 396.5 Mb of the assembled genome is placed in the linkage map ( supplementary table S2 , Supplementary Material online).

Table 1

Genome Assembly Statistics

Genome  
Size (1n) 915 Mb 
Karyotype 1n = 24 
GC content 39% 
Assembly  
Size in scaffolds > 500bp 654 Mb 
Number of scaffolds >500bp 40,758 
Number of scaffolds > 10kb 7,929 
Base pairs placed in linkage map 396.7 Mb 
Number of scaffolds placed in linkage map 4871 
N50 scaffold 111.5 kb 
N50 contig 16.1 kb 
Annotation  
Coding loci 27,328 
Noncoding loci 1,376 
Genome  
Size (1n) 915 Mb 
Karyotype 1n = 24 
GC content 39% 
Assembly  
Size in scaffolds > 500bp 654 Mb 
Number of scaffolds >500bp 40,758 
Number of scaffolds > 10kb 7,929 
Base pairs placed in linkage map 396.7 Mb 
Number of scaffolds placed in linkage map 4871 
N50 scaffold 111.5 kb 
N50 contig 16.1 kb 
Annotation  
Coding loci 27,328 
Noncoding loci 1,376 

Evidence-guided annotation of the genome assembly resulted in 27,328 protein-coding genes, 26 ribosomal RNAs, 536 transfer RNAs, and 814 other noncoding RNAs including microRNAs and small nucleolar RNAs ( supplementary tables S3 and S4 , Supplementary Material online). Of the protein-coding genes, 84% share sequence similarity to 16,919 proteins in the SWISS-PROT database ( e- value 10 10 ) ( Boeckmann et al. 2003 ). The mitochondrial genome is absent from our assembly due to the high raw read coverage.

The gene regions for K. marmoratus in this assembly are relatively well assembled. Of the highly conserved core eukaryotic genes determined by CEGMA, 208 complete proteins and an additional 28 partial proteins were identified in the genome assembly, which, in total, correspond to a 95% of CEGMA proteins ( supplementary table S5 , Supplementary Material online). Moreover, 76.5% complete and 10.2% fragmented single-copy vertebrate orthologs were identified using Benchmarking Universal Single-Copy Orthologs (BUSCO) ( supplementary table S6 , Supplementary Material online). This completeness of BUSCO genes is similar to other widely used and well-assembled fish genomes ( supplementary table S7 , Supplementary Material online) To support the completeness of the genome, one of the short-insert libraries used for the genome assembly maps back to the reference genome with a mapping rate of 98.7%. Moreover, 90.5% of transcriptome reads from an unrelated study ( Mesak et al. 2015 ), map to the reference genome. More than 30% of the unmapped reads map to the mitochondrial genome sequence available on GenBank (NC_003290, Lee et al. 2001 ).

Comparison to Sister Species, K. hermaphroditus

To compare K. marmoratus to K. hermaphroditus , we sequenced a lab-reared K. hermaphroditus individual to approximately 7-fold coverage and compared the resulting sequence data to our genome assembly. For this analysis, we appended the existing mitochondrial genome to our genome assembly. The reads map to the reference assembly with a 97.8% mapping rate. Of sites that met our coverage threshold for comparison between the reference and K. hermaphroditus , 0.4% (1,739,544/421,508,114) were homozygous for an alternate allele and 0.1% (432,064/421,508,114) were heterozygous for an alternate allele. All other alleles were identical between K. marmoratus and K. hermaphroditus .

As K. marmoratus and K. hermaphroditus are sister species, we were interested in characterizing loci that were highly differentiated between the two species. We identified highly differentiated genes and overrepresented gene ontology (GO) terms for the comparison. We identified 23,598 genes that had at least one site where the RHL individual was homozygous for one allele and the K. hermaphroditus individual was homozygous for another allele. We ranked genes based on the number of differences between K. marmoratus and K. hermaphroditus and considered any genes that were in the top 1% of distribution as highly differentiated (241 genes). Of the 241 highly differentiated genes, 218 had a significant hit in the BLAST search and, of those, 204 had a human homolog and were associated with at least one biological process GO (see Materials and Methods) term. Our resulting list included 1,091 GO terms. We tested for enrichment of biological process, molecular function, and cellular component GO terms in our highly differentiated genes using Webgestalt ( Wang et al. 2013 ). The most highly enriched terms in biological process were nervous system development (GO:0007399), multicellular organismal process (GO:0032501), and single-multicellular organism process (GO:0044707). For molecular function, transmembrane signaling receptor activity (GO:0004888), receptor activity (GO:0004872), and signaling receptor activity (GO:0038023), were the most significant terms. Finally, terms associated with cellular component that were most enriched were membrane related (GO:0031224, GO:0016021, GO:0044425).

Sex Determination Loci

Due to the environmental sex determination (hermaphrodite vs. primary male) and sexual plasticity (hermaphrodite to secondary male transition) ( fig. 1 ), we focused on annotating and comparing known sex determination loci from other systems. We identified major known sex-determining loci and report their gene names along with the scaffold position and linkage group, when applicable ( supplementary table S8 , Supplementary Material online). We identified two sox9 genes, whose homologs have been identified in zebrafish and appear to have arisen during the teleost whole-genome duplication event ( Chiang et al. 2001 ). None of the sex determining loci was in the set of highly differentiated loci, which was expected because both species are hermaphrodites. We analyzed the sex determining genes for elevated rates of nonsynonymous to synonymous substitutions (d N /d S ) using both a sites model ( Yang 2007 ) and a branch-sites model ( Yang and dos Reis 2011 ) with the branch leading to K. marmoratus and K. hermaphroditus as the foreground branch. None of the loci had significant evidence for positive selection using either of these approaches with P < 0.01. Some of genes that have been implicated in sex determination have complex roles in development and we speculate that gene expression changes rather than structural changes may be driving differences in these species.

Comparison to Other Fish Species

We compared our protein-coding gene annotation set with annotated proteins from zebrafish ( Danio rerio ) ( Howe et al. 2013 ), stickleback ( Gasterosteus aculeatus ) ( Jones et al. 2012 ), medaka ( Oryzias latipes ) ( Kasahara et al. 2007 ), fugu ( Takifugu rubripes ) ( Aparicio et al. 2002 ), Amazon molly ( Poecilia formosa ) (W. Warren and The Genome Institute, Washington University School of Medicine, GenBank Assembly ID GCA_000485575.1), mummichog ( Fundulus heteroclitus ) (GenBank Assembly ID GCA_000826765.1), and turquoise killifish ( Nothobranchius furzeri ) ( Dario et al. 2015 ) ( fig. 2 ). In a comparison among the eight species, 4,827 genes were unique to mangrove rivulus. It is important to note that defining unique genes is a difficult endeavor, especially in teleost fish, and these are preliminary results. Of the 4,827 genes, 2,465 of them (∼51%) have a BLAST hit in the SwissProt database and human orthologs exist for 2,213 of them (∼90%) and can be used for GO enrichment analyses. The enriched GO terms include many terms relating to the nervous system and ion channels.

Materials and Methods

Sample Collection and DNA/RNA Extraction

The K. marmoratus genome was sequenced and assembled using DNA from a lineage RHL derived from San Salvador, Bahamas; the progenitor was caught in the field by D. Scott Taylor in 1997 and the animals used here were 10–11 generations removed from the field. DNA was extracted from muscle tissue from multiple K. marmoratus RHL specimens from the Earley laboratory. Genetic homogeneity of the RHL lineage was verified by genotyping the specimens at 32 microsatellite loci developed for K. marmoratus ( Mackiewicz, Tatarenkov, Perry, et al. 2006 ). For each locus, one polymerase chain reaction (PCR) primer in each pair was labeled with a fluorescent dye (HEX, 6-FAM, or NED) and DNA was amplified in several multiplex reactions, as described previously ( Tatarenkov et al. 2012 ). Multiplexed PCR products were diluted 10–20 fold and 1 µl of the diluted product was mixed with 9.6 µl of deionized formamide and 0.4 µl size standard GS500 (ROX labeled; Applied Biosystems), denatured for 4 min at 95 °C, and electrophoresed on an GA 3100 using 50 cm capillaries filled with Pop4 (Applied Biosystems). Alleles were scored using Genemapper 4.0 (Applied Biosystems). All specimens were fully homozygous, genetically identical, and matched microsatellite profile of the RHL lineage described in Tatarenkov et al. (2010) . DNA was extracted using the Gentra Puregene Tissue Kit (Qiagen) according to the manufacturer’s instructions. Purified DNA was sheared using a Covaris E220 (Woburn, MA) (duty cycle 10, intensity 5, cycles/burst 200, time 180 s) to approximately 400 bp. Sheared genomic DNA was gel purified and used as input for sequencing library preparation. Sheared genomic DNA was end repaired with NEBNext end-repair kit (E6050L), and A-tailing was accomplished using Taq polymerase. Ultrapure ligase (L603-HC-L) from Enzymatics and iProof (BioRad) were used for ligation of the Illumina paired-end adapters (PE‐102-1003) and amplification, respectively. Agencourt Ampure XP beads were used for clean-up at each step. Three libraries of the RHL lineage were sequenced using Illumina HiSeq2000 sequencing technology with 2 × 101 bp paired-end reads at the Stanford Center for Genomics and Personalized Medicine (Raw reads have been deposited under NCBI BioProject PRJNA290522). Reads were filtered and trimmed using Trim Galore! v.0.2.2 ( Krueger 2014 ), retaining reads with average Phred quality greater than 20 and at least 50 bp in length. An additional long mate-pair library was generated in collaboration with Life Technologies (now Thermo Fisher Scientific) for the IonTorrent PGM following manufacturer’s protocols for mate-pair library construction with a target insert size of over 15 kb. Size selection for the appropriate insert size was performed using pulse field gel electrophoresis.

RNA was isolated from liver, gonad, and embryo tissue for a lab-reared specimen of RHL. RNA was isolated by pulverizing 50–100 mg of tissue frozen in liquid nitrogen with a Covaris Cryoprep at setting 3. RNA was then extracted with Qiagen’s RNeasy Plus mini kit. Total RNA was enriched for non-rRNA using either Ribo-zero (Epicenter) for liver and gonad NEBNext® Poly(A) mRNA Magnetic Isolation module. Enriched RNA was fragmented to an average size of 400 nt using NEB mRNA Fragmentation Module by incubation at 94 °C for 4 min. Fragmented enriched RNA was purified using Agencourt RNAClean XP beads. First-strand and second-stranded cDNA was synthesized and the reaction was purified with Agencourt Ampure XP beads. Double-stranded cDNA was used as input for Illumina sequencing library preparation using the NEBNext end-repair kit, A-tailing with Taq polymerase, ligation with barcoded adapters, and amplification with Kapa Library Amplification Readymix or as part of the NEBNext® Ultra Directional RNA Library Prep Kit.

DNA was extracted from liver tissue from a lab-reared specimen of K. hermaphroditus (strain GITMO), also from the Earley laboratory. A small-insert Illumina sequencing library was generated as described above for RHL and sequenced on an Illumina HiSeq2000 with 2 × 101 bp paired-end reads at the Stanford Center for Genomics and Personalized Medicine (Raw reads have been deposited under NCBI BioProject PRJNA290522).

Genome Size Estimation

Genome size was estimated from the raw reads using the k-mer frequency distribution for k = 17 and k = 23, as in Li et al. (2010) . K-mer frequencies were counted using JELLYFISH ( Marcais and Kingsford 2011 ).

Assembly Strategy

De novo genome assembly was accomplished using a multistep process. Assembly of the genome using the short read Illumina data included error correction, preprocessing, and assembly with the String Graph Assembler (SGA) version 0.10.1 ( Simpson and Durbin 2012 ). Scaffolding with the mate pair data was accomplished using SSPACE Basic v2.0 ( Boetzer et al. 2011 ). The final draft assembly contains scaffolds at least 500 bp. The final assembly is deposited under NCBI BioProject PRJNA290522. Genome assembly statistics were computed using a Perl script, provided by Assemblathon 2 ( Bradnam et al. 2013 ), and CEGMA v.2.5 ( Parra et al. 2007 ). CEGMA required dependencies included geneid v.1.4.4 ( Blanco et al. 2002 ), genewise v.2.4.1 ( Birney et al. 2004 ), and HMMER 3.1b2 ( Mistry et al. 2013 ). CEGMA was used to determine the percent of a defined core set of 248 highly conserved eukaryotic genes is present in the assembly. BUSCO (version 1.1b1) was also used to assess assembly completeness, using the lineage vertebrata. BUSCO dependencies included HMMER 3.1b2 ( Mistry et al. 2013 ) and augustus.2.5.5 ( Stanke et al. 2008 ).

To perform higher-order scaffolding, the linkage map from ( Kanamori et al. 2016 ) was used. Markers were mapped in fasta format to the genome assembly using Burrows–Wheeler Aligner (BWA; Li and Durbin 2009 ). Information regarding the scaffold and the mapped markers was parsed from alignment file. Markers with multiple mappings were discarded. Information about orientation is not available for the linkage map, therefore the mapping of scaffolds to linkage groups is included in supplementary table S2 , Supplementary Material online.

Annotation Strategy

The MAKER2 annotation pipeline ( Holt and Yandell 2011 ) was utilized for gene annotation. Homo sapiens UniProt ( UniProt Consortium 2012 ) and Austrofundulus limnaeus Annotation Release 100 (GenBank GCF_001266775.1) protein databases were used for protein homology data. RepeatMasker was used to mask low complexity regions in the genome assembly ( Smit et al. 2013–2015 ; supplementary table S9 , Supplementary Material online). Filtered RNA-sequencing data was mapped to the genome using Bowtie ( Langmead et al. 2009 ), assembled into transcriptomes using Cufflinks ( Roberts et al. 2011 ) and used as EST evidence in MAKER2. Ab initio gene prediction was accomplished using two rounds of training with SNAP ( Korf 2004 ). Annotation of the predicted protein coding genes was accomplished with homology searching using BLASTP ( Altschul et al. 1990 ) with a threshold cutoff e- value of 10 10 against the Swiss-Prot database ( Boeckmann et al. 2003 ). Ribosomal RNAs and other noncoding RNAs were annotated using Rfam ( Nawrocki et al. 2015 ); tRNAs were additional annotated using tRNAscan-SE 1.3.1 (options -H and -y) ( Lowe and Eddy 1997 ) and Aragorn 1.2.34 (option -w -t -i116 -l -d) (Laslett and Canback 2004 ).

Comparison to Other Data Sets

Transcriptome reads from Mesak et al (2015) were downloaded from the Short Read Archive (SRR2001227) and mapped to the reference genome using BWA ( Li and Durbin 2009 ). Raw reads generated in this study, from RHL and K. hermaphroditus , were separately mapped to the reference genome using BWA ( Li and Durbin 2009 ). File manipulation and summary statistics were generated using SAMtools ( Li et al. 2009 ) and BamTools ( Barnett et al. 2011 ). The mitochondrial reference sequence (Genbank, NC_003290) was included for the K. hermaphroditus mapping. Genotypes were called separately for each individual using the UnifiedGenotyper tool in the Genome Analysis Toolkit (GATK) (v. 3.5) with the EMIT_ALL_CONFIDENT_SITES option. ( McKenna et al. 2010 ) Individual vcf files were then filtered following the GATK recommended best practices ( DePristo et al. 2011 ; Van der Auwera et al. 2013 ). Vcfs were filtered using vcftools (v. 0.1.12b) such that only sites with at least 6× coverage (–minDP 6) were kept for each individual ( Danecek et al. 2011 ). The individual vcf files were combined using the CombineVariants tool in GATK ( McKenna et al. 2010 ) and only sites with genotypes in both individuals were retained with –max-missing 1.0 in vcftools ( Danecek et al. 2011 ). Sites where the two individuals were homozygous for different alleles were pulled out for further analysis. We identified which of these sites were located in genes using the intersectbed utility in BEDtools (v. 2.17.0) ( Quinlan and Hall 2010 ).

Sex Determination Loci

To identify genes implicated in sex determination processes, we queried our annotation set for the orthologous loci and report additional details about linkage group and relative position across contigs.

GO Enrichment

To characterize putatively differentiated genes, we used bioDBnet ( Mudunuri et al. 2009 ) to identify biological process GO terms that were associated with the highly differentiated genes. We also tested for enrichment of biological process, molecular function, and cellular component GO terms in our highly differentiated genes. BioDBnet ( Mudunuri et al. 2009 ) was used to convert gene names into human Entrez IDs. We used Webgestalt ( Wang et al. 2013 ) to test for enrichment of GO terms compared to a reference set composed of human orthologs of the annotated genes in the K. marmoratus genome. Significantly overrepresented GO terms were identified at a false discovery rate of 0.01 ( Benjamini and Hochberg 1995 ). At least two genes had to be associated with a GO term for it to be considered enriched.

To determine sets of orthologous genes, we compared protein data sets among closely related fish species. We included protein data from D. rerio (43,153), G. aculeatus (27,576), O. latipes (24,674), T. rubripes (47,841), P. formosa (30,898), F. heteroclitus (33,705), and N. furzeri (30,028) in our comparative analysis. Clusters of orthologous genes were identified using OrthoFinder ( Emms and Kelly 2015 ) with MCL ( Enright et al. 2002 ).

Supplementary Material

Supplementary tables S1–S9 are available at Genome Biology and Evolution online ( http://www.gbe.oxfordjournals.org/ ).

Acknowledgments

This work was in part supported by a National Institutes of Health National Research Service Award postdoctoral fellowship [GM087069 to J.L.K.]. The authors thank Craig Cummings, Elizabeth Levandowsky, Minita Shah from Life Technologies for their assistance in library preparation and sequencing. The authors would like to thank Jared Simpson for useful feedback on genome assembly with SGA.

Literature Cited

Abzhanov
A
, et al.   .
2008
.
Are we there yet? Tracking the development of new model systems
.
Trends Genet.
 
24
:
353
360
.
Altschul
SF
Gish
W
Miller
W
Myers
EW
Lipman
DJ.
1990
.
Basic local alignment search tool
.
J Mol Biol.
 
215
:
403
410
.
Aparicio
S
, et al.   .
2002
.
Whole-Genome Shotgun Assembly and Analysis of the Genome of Fugu rubripes
.
Science
 
297
:
1301
1310
.
Ashley-Ross
MA
Perlman
BM
Gibb
AC
Long
JH
Jr.
2014
.
Jumping sans legs: does elastic energy storage by the vertebral column power terrestrial jumps in bony fishes?
Zoology (Jena)
 
117
:
7
18
.
Barnett
DW
Garrison
EK
Quinlan
AR
Stromberg
MP
Marth
GT.
2011
.
BamTools: a C ++ API and toolkit for analyzing and managing BAM files
.
Bioinformatics
 
27
:
1691
1692
.
Benjamini
Y
Hochberg
Y.
1995
.
Controlling the false discovery rate: a practical and powerful approach to multiple testing
.
J Roy Stat Soc Ser B (Methodol)
  .
57
:
289
300
.
Birney
E
Clamp
M
Durbin
R.
2004
.
Genewise and genomewise
.
Genome Res.
 
14
:
988
995
.
Blanco
E
Parra
G
Guigó
R.
2002
. Using geneid to Identify Genes . In:
Baxevanis
AD
Davison
DB
, editors.
Current protocols in bioinformatics
  . Volume
1
, Unit 4.3.
New York
:
John Wiley & Sons Inc
.
18:4.3:4.3.1–4.3.28.
Boeckmann
B
, et al.   .
2003
.
The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003
.
Nucleic Acids Res.
 
31
:
365
370
.
Boetzer
M
Henkel
CV
Jansen
HJ
Butler
D
Pirovano
W.
2011
.
Scaffolding pre-assembled contigs using SSPACE
.
Bioinformatics
 
27
:
578
579
.
Bradnam
KR
, et al.   .
2013
.
Assemblathon 2: evaluating de novo methods of genome assembly in three vertebrate species
.
Gigascience
 
2
:
10.
Chiang
EFL
, et al.   .
2001
.
Two Sox9 genes on duplicated zebrafish chromosomes: expression of similar transcription activators in distinct sites
.
Dev Biol.
 
231
:
149
163
.
Cole
KS
Noakes
LGD.
1997
.
Gonadal development and sexual allocation in mangrove killifish, Rivulus marmoratus (Pisces: Atherinomorpha)
.
Copeia
 
1997
:
596
600
.
Costa
WJEM.
2006
.
Redescription of Kryptolebias ocellatus (Hensel) and K. caudomarginatus (Seegers)(Teleostei: Cyprinodontiformes: Rivulidae), two killifishes from mangroves of south-eastern Brazil J. Ichthyol
.
Aquatic Biol.
 
11
:
5
12
.
Costa
WJEM.
2011
.
Identity of Rivulus ocellatus and a new name for a hermaphroditic species of Kryptolebias from south-eastern Brazil (Cyprinodontiformes: Rivulidae)
.
Ichthyol Explor Freshw
  .
22
:
185
192
.
Danecek
P
, et al.   .
2011
.
The variant call format and VCFtools
.
Bioinformatics
 
27
:
2156
2158
.
Davis
WP.
1986
.
The role of Rivulus marmoratus in research on aquatic pollutants
.
J Am Killifish Assoc
  .
19
:
70
80
.
DePristo
MA
, et al.   .
2011
.
A framework for variation discovery and genotyping using next-generation DNA sequencing data
.
Nat Genet.
 
43
:
491
498
.
Earley
RL
, et al.   .
2012
.
The evolution of complex phenotypes in mangrove rivulus ( Kryptolebias marmoratus ): a prospective
.
Integr Comp Biol.
 
52
:
814
827
.
Earley
RL
Hsu
Y.
2008
.
Reciprocity between endocrine state and contest behavior in the killifish, Kryptolebias marmoratus
.
Horm Behav
  .
53
:
442
451
.
Earley
RL
Lu
CK
Lee
IH
Wong
SC
Hsu
Y.
2013
.
Winner and loser effects are modulated by hormonal states
.
Front Zool
  .
10
:
6.
Edenbrow
M
Croft
DP.
2011
.
Behavioural types and life history strategies during ontogeny in the mangrove killifish, Kryptolebias marmoratus
.
Anim Behav
  .
82
:
731
741
.
Edenbrow
M
Croft
DP.
2012
.
Kin and familiarity influence association preferences and aggression in the mangrove killifish Kryptolebias marmoratus
.
J Fish Biol.
 
80
:
503
518
.
Ellison
A
, et al.   .
2015
.
Epigenetic regulation of sex ratios may explain natural variation in self-fertilization rates
.
Proc Biol Sci.
 
282
. doi: 10.1098/rspb.2015.1900.
Emms
DM
Kelly
S.
2015
.
OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy
.
Genome Biol.
 
16
:
157.
Enright
AJ
Van Dongen
S
Ouzounis
CA.
2002
.
An efficient algorithm for large-scale detection of protein families
.
Nucleic Acids Res.
 
30
:
1575
1584
.
Farmer
JL
Orlando
EF.
2012
.
Creating females? Developmental effects of 17alpha-ethynylestradiol on the mangrove rivulus' ovotestis
.
Integr Comp Biol.
 
52
:
769
780
.
Furness
AI
Tatarenkov
A
Avise
JC.
2015
.
A genetic test for whether pairs of hermaphrodites can cross-fertilize in a selfing killifish
.
J Hered
  .
106
:
749
752
.
Garcia
MJ
, et al.   .
2016
.
Phenotypic differences between the sexes in the sexually plastic mangrove rivulus fish ( Kryptolebias marmoratus )
.
J Exper Biol.
 
219
:
988
997
.
Garcia
MJ
Williams
J
Sinderman
B
Earley
RL.
2015
.
Ready for a fight? The physiological effects of detecting an opponent's pheromone cues prior to a contest
.
Physiol Behav
  .
149
:
1
7
.
Gibson
DJ
Sylvester
EVA
Turko
AJ
Tattersall
GJ
Wright
PA.
2015
.
Out of the frying pan into the air—emersion behaviour and evaporative heat loss in an amphibious mangrove fish ( Kryptolebias marmoratus )
.
Biol Lett.
 
11
. doi: 10.1098/rsbl.2015.0689.
Harrington
RW.
1975
. Sex determination and differentiation among uniparental homozygotes of the hermaphroditic fish Rivulus marmoratus (Cyprinodontidae: Atheriniformes). In:
Reinboth
R
, editor.
Intersexuality in the animal kingdom
  .
Berlin, Heidelberg
:
Springer Berlin Heidelberg
. p.
249
262
.
Harrington
RW
Jr.
1961
.
Oviparous hermaphroditic fish with internal self-fertilization
.
Science
 
134
:
1749
1750
.
Harrington
RW
Jr.
1967
.
Environmentally controlled induction of primary male gonochorists from eggs of the self-fertilizing hermaphroditic fish, Rivulus marmoratus
.
Biol Bull.
 
132
:
174
199
.
Harrington
RW
Jr.
1968
.
Delimitation of the thermolabile phenocritical period of sex determination and differentiation in the ontogeny of the normally hermaphroditic fish Rivulus marmoratus Poey
.
Physiol Zool
  .
41
:
447
460
.
Harrington
RW
Jr.
1971
.
How ecological and genetic factors interact to determine when self-fertilizing hermaphrodites of Rivulus marmoratus change into functional secondary males, with a reappraisal of the modes of intersexuality among fishes
.
Copeia
 
1971
:
389
432
.
Harrington
RW
Jr
Kallman
KD.
1968
.
The homozygosity of clones of the self-fertilizing hermaphroditic fish Rivulus marmoratus Poey (Cyprinodontidae, Atheriniformes)
.
Am Nat
  .
102
:
337
343
.
Holt
C
Yandell
M.
2011
.
MAKER2: an annotation pipeline and genome-database management tool for second-generation genome projects
.
BMC Bioinformatics
 
12
:
491.
Howe
K
, et al.   .
2013
.
The zebrafish reference genome sequence and its relationship to the human genome
.
Nature
 
496
:
498
503
.
Hsu
Y
Lee
S-P
Chen
M-H
Yang
S-Y
Cheng
K-C.
2008
.
Switching assessment strategy during a contest: fighting in killifish Kryptolebias marmoratus
.
Animal Behav
  .
75
:
1641
1649
.
Johnson
EL
Weinersmith
KL
Earley
RL.
2016
.
Changes in reproductive physiology of mangrove rivulus Kryptolebias marmoratus following exposure to environmentally relevant doses of ethinyl oestradiol
.
J Fish Biol.
 
88
:
774
786
.
Jones
FC
, et al.   .
2012
.
The genomic basis of adaptive evolution in threespine sticklebacks
.
Nature
 
484
:
55
61
.
Kallman
KD
Harrington
RW.
1964
.
Evidence for the existence of homozygous clones in the self-fertilizing hermaphroditic teleost Rivulus marmoratus (Poey)
.
Biol Bull.
 
126
:
101
114
.
Kanamori
A
, et al.   .
2006
.
Methyltestosterone efficiently induces male development in the self-fertilizing hermaphrodite fish, Kryptolebias marmoratus
.
Genesis
 
44
:
495
503
.
Kanamori
A
, et al.   .
2016
.
A genetic map for the only self-fertilizing vertebrate
.
G3
 
6
:
1095
1106
.
Kasahara
M
, et al.   .
2007
.
The medaka draft genome and insights into vertebrate genome evolution
.
Nature
 
447
:
714
719
.
Korf
I.
2004
.
Gene finding in novel genomes
.
BMC Bioinformatics
 
5
:
59.
Kristensen
I.
1970
.
Competition in three cyprinodont fish species in the Netherlands
  .
Antilles
:
M. Nijhoff.
Krueger
F.
2014
. TrimGalore! Available from: http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/ .
Langmead
B
Trapnell
C
Pop
M
Salzberg
SL.
2009
.
Ultrafast and memory-efficient alignment of short DNA sequences to the human genome
.
Genome Biol.
 
10
:
R25.
Laslett
D
Canback
B.
2004
.
ARAGORN, a program to detect tRNA genes and tmRNA genes in nucleotide sequences
.
Nucleic Acids Res.
 
32
:
11
16
.
Laughlin
TF
Lubinski
BA
Park
EH
Taylor
DS
Turner
BJ.
1995
.
Clonal stability and mutation in the self-fertilizing hermaphroditic fish, Rivulus marmoratus
.
J Hered
  .
86
:
399
402
.
Lee
JS
, et al.   .
2001
.
The complete DNA sequence of the mitochondrial genome of the self-fertilizing fish Rivulus marmoratus (Cyprinodontiformes, Rivulidae) and the first description of duplication of a control region in fish
.
Gene
 
280
:
1
7
.
Lee
JS
Raisuddin
S
Schlenk
D.
2008
.
Kryptolebias marmoratus (Poey, 1880): a potential model species for molecular carcinogenesis and ecotoxicogenomics
.
J Fish Biol.
 
72
:
1871
1889
.
Lee
YM
, et al.   .
2007
.
Mining of biomarker genes from expressed sequence tags and differential display reverse transcriptase-polymerase chain reaction in the self-fertilizing fish, Kryptolebias marmoratus and their expression patterns in response to exposure to an endocrine-disrupting alkylphenol, bisphenol A
.
Mol Cells
  .
23
:
287
303
.
Lee
YM
Seo
JS
Kim
IC
Yoon
YD
Lee
JS.
2006
.
Endocrine disrupting chemicals (bisphenol A, 4-nonylphenol, 4-tert-octylphenol) modulate expression of two distinct cytochrome P450 aromatase genes differently in gender types of the hermaphroditic fish Rivulus marmoratus
.
Biochem Biophys Res Commun.
 
345
:
894
903
.
Lehner
B.
2013
.
Genotype to phenotype: lessons from model organisms for human genetics
.
Nat Rev Genet.
 
14
:
168
178
.
Li
H
Durbin
R.
2009
.
Fast and accurate short read alignment with Burrows-Wheeler transform
.
Bioinformatics
 
25
:
1754
1760
.
Li
H
, et al.   .
2009
.
The Sequence Alignment/Map format and SAMtools
.
Bioinformatics
 
25
:
2078
2079
.
Li
R
, et al.   .
2010
.
The sequence and de novo assembly of the giant panda genome
.
Nature
 
463
:
311
317
.
Lin
HC
Dunson
WA.
1999
.
Phenotypic plasticity in the growth of the self-fertilizing hermaphroditic fish Rivulus marmoratus
.
J Fish Biol.
 
54
:
250
266
.
Lowe
TM
Eddy
SR.
1997
.
tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence
.
Nucleic Acids Res.
 
25
:
955
964
.
Lubinski
BA
Davis
WP
Taylor
DS
Turner
BJ.
1995
.
Outcrossing in a natural population of self-fertilizing hermaphroditic fish
.
J Hered
  .
86
:
469
473
.
Mackay
TF.
2014
.
Epistasis and quantitative traits: using model organisms to study gene-gene interactions
.
Nat Rev Genet.
 
15
:
22
33
.
Mackay
TF
Stone
EA
Ayroles
JF.
2009
.
The genetics of quantitative traits: challenges and prospects
.
Nat Rev Genet.
 
10
:
565
577
.
Mackiewicz
M
Tatarenkov
A
Perry
A
, et al.   .
2006
.
Microsatellite documentation of male-mediated outcrossing between inbred laboratory strains of the self-fertilizing mangrove killifish ( Kryptolebias marmoratus )
.
J Hered
  .
97
:
508
513
.
Mackiewicz
M
Tatarenkov
A
Taylor
DS
Turner
BJ
Avise
JC.
2006
.
Extensive outcrossing and androdioecy in a vertebrate species that otherwise reproduces as a self-fertilizing hermaphrodite
.
Proc Natl Acad Sci U S A.
 
103
:
9924
9928
.
Mackiewicz
M
Tatarenkov
A
Turner
BJ
Avise
JC.
2006
.
A mixed-mating strategy in a hermaphroditic vertebrate
.
Proc Biol Sci.
 
273
:
2449
2452
.
Marcais
G
Kingsford
C.
2011
.
A fast, lock-free approach for efficient parallel counting of occurrences of k-mers
.
Bioinformatics
 
27
:
764
770
.
McKenna
A
, et al.   .
2010
.
The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data
.
Genome Res.
 
20
:
1297
1303
.
Mesak
F
Tatarenkov
A
Avise
JC.
2015
.
Transcriptomics of diapause in an isogenic self-fertilizing vertebrate
.
BMC Genomics
 
16
:
989.
Mistry
J
Finn
RD
Eddy
SR
Bateman
A
Punta
M.
2013
.
Challenges in homology search: HMMER3 and convergent evolution of coiled-coil regions
.
Nucleic Acids Res.
 
41
:
e121.
Moore
G
, et al.   .
2012
.
Establishing Developmental Genetics in a Self-Fertilizing Fish ( Krytolebias marmoratus )
.
Integ Comp Biol.
 
52
:
781
791
.
Mourabit
S
Edenbrow
M
Croft
DP
Kudoh
T.
2011
.
Embryonic development of the self-fertilizing mangrove killifish Kryptolebias marmoratus
.
Dev Dyn
  .
240
:
1694
1704
.
Mourabit
S
Kudoh
T.
2012
.
Manipulation and imaging of Kryptolebias marmoratus embryos
.
Integr Comp Biol.
 
52
:
761
768
.
Mudunuri
U
Che
A
Yi
M
Stephens
RM.
2009
.
bioDBnet: the biological database network
.
Bioinformatics
 
25
:
555
556
.
Nakamura
Y
Suga
K
Sakakura
Y
Sakamoto
T
Hagiwara
A.
2008
.
Genetic and growth differences in the outcrossings between two clonal strains of the self-fertilizing mangrove killifish
.
Can J Zool
  .
86
:
976
982
.
Nawrocki
EP
, et al.   .
2015
.
Rfam 12.0: updates to the RNA families database
.
Nucleic Acids Res.
 
43
:
D130
D137
.
Orlando
EF
Katsu
Y
Miyagawa
S
Iguchi
T.
2006
.
Cloning and differential expression of estrogen receptor and aromatase genes in the self-fertilizing hermaphrodite and male mangrove rivulus, Kryptolebias marmoratus
.
J Mol Endocrinol
  .
37
:
353
365
.
Park
CB
, et al.   .
2013
.
Transient effects of methyltestosterone injection on different reproductive parameters of the hermaphrodite fish Kryptolebias marmoratus
.
Ecotoxicology
 
22
:
1145
1154
.
Parra
G
Bradnam
K
Korf
I.
2007
.
CEGMA: a pipeline to accurately annotate core genes in eukaryotic genomes
.
Bioinformatics
 
23
:
1061
1067
.
Pronko
AJ
Perlman
BM
Ashley-Ross
MA.
2013
.
Launches, squiggles and pounces, oh my! The water–land transition in mangrove rivulus ( Kryptolebias marmoratus )
.
J Exper Biol.
 
216
:
3988
3995
.
Quinlan
AR
Hall
IM.
2010
.
BEDTools: a flexible suite of utilities for comparing genomic features
.
Bioinformatics
 
26
:
841
842
.
Rankin
CH.
2015
.
A review of transgenerational epigenetics for RNAi, longevity, germline maintenance and olfactory imprinting in Caenorhabditis elegans
.
J Exp Biol.
 
218
:
41
49
.
Rhee
J-S
, et al.   .
2008
.
Gonadotropin-releasing hormone receptor (GnRHR) gene expression is differently modulated in gender types of the hermaphroditic fish Kryptolebias marmoratus by endocrine disrupting chemicals
.
Comp Biochem Physiol C Toxicol Pharmacol
  .
147
:
357
365
.
Rhee
JS
, et al.   .
2009
.
Endocrine disruptors modulate expression of hepatic choriogenin genes in the hermaphroditic fish, Kryptolebias marmoratus
.
Comp Biochem Physiol C Toxicol Pharmacol
  .
150
:
170
178
.
Rhee
JS
, et al.   .
2010
.
Bisphenol A modulates expression of gonadotropin subunit genes in the hermaphroditic fish, Kryptolebias marmoratus
.
Comp Biochem Physiol C Toxicol Pharmacol
  .
152
:
456
466
.
Rhee
JS
, et al.   .
2011
.
Bisphenol A modulates expression of sex differentiation genes in the self-fertilizing fish, Kryptolebias marmoratus
.
Aquat Toxicol
  .
104
:
218
229
.
Rhee
JS
Lee
JS.
2014
.
Whole genome data for omics-based research on the self-fertilizing fish Kryptolebias marmoratus
.
Mar Pollut Bull.
 
85
:
532
541
.
Rhee
J-S
Lee
Y-M
Seo
JS
Han
J
Lee
J-S.
2009
.
Expression of gonadotropin α, follicle-stimulating hormone β, and luteinizing hormone β genes of the hermaphroditic fish Kryptolebias marmoratus exposed to octylphenol, 17β estradiol, and tamoxifen
.
Ann N Y Acad Sci.
 
1163
:
508
511
.
Roberts
A
Pimentel
H
Trapnell
C
Pachter
L.
2011
.
Identification of novel transcripts in annotated genomes using RNA-Seq
.
Bioinformatics
 
27
:
2325
2329
.
Robertson
C
Turko
AJ
Jonz
MG
Wright
PA.
2015
.
Hypercapnia and low pH induce neuroepithelial cell proliferation and emersion behaviour in the amphibious fish Kryptolebias marmoratus
.
J Exper Biol.
 
218(Pt 19)
:
2987
2990
.
Scheel
JJ.
1972
.
Rivuline karyotypes and their evolution (rivulinae, cyprinodontidae, pisces)
.
J Zool Syst Evol Res.
 
10
:
180
209
.
Simpson
JT
Durbin
R.
2012
.
Efficient de novo assembly of large genomes using compressed data structures
.
Genome Res.
 
22
:
549
556
.
Smit
A
Hubley
R
Green
P.
2013–2015
. RepeatMasker Open-4.0. Available from: http://www.repeatmasker.org .
Stanke
M
Diekhans
M
Baertsch
R
Haussler
D.
2008
.
Using native and syntenically mapped cDNA alignments to improve de novo gene finding
.
Bioinformatics
 
24
:
637
644
.
Sucar
S
Moore
GL
Ard
ME
Ring
BC.
2016
.
A simultaneous genetic screen for zygotic and sterile mutants in a hermaphroditic vertebrate ( Kryptolebias marmoratus )
.
G3 (Bethesda)
 
6
:
1107
1119
.
Tatarenkov
A
Earley
RL
Taylor
DS
Avise
JC.
2012
.
Microevolutionary distribution of isogenicity in a self-fertilizing fish ( Kryptolebias marmoratus ) in the Florida keys
.
Integr Comp Biol.
 
52
:
743
752
.
Tatarenkov
A
, et al.   .
2007
.
Strong population structure despite evidence of recent migration in a selfing hermaphroditic vertebrate, the mangrove killifish ( Kryptolebias marmoratus )
.
Mol Ecol
  .
16
:
2701
2711
.
Tatarenkov
A
, et al.   .
2015
.
Genetic subdivision and variation in selfing rates among Central American populations of the mangrove rivulus, Kryptolebias marmoratus
.
J Hered
  .
106
:
276
284
.
Tatarenkov
A
Ring
BC
Elder
JF
Bechler
DL
Avise
JC.
2010
.
Genetic composition of laboratory stocks of the self-fertilizing fish Kryptolebias marmoratus : a valuable resource for experimental research
.
PLoS One
 
5
:
e12863.
Taylor
DS.
2001
.
Physical variability and fluctuating asymmetry in heterozygous and homozygous populations of Rivulus marmoratus
.
Can J Zool
  .
79
:
766
778
.
Taylor
DS.
2012
.
Twenty-four years in the mud: what have we learned about the natural history and ecology of the mangrove rivulus, Kryptolebias marmoratus ?
Integr Comp Biol.
 
52
:
724
736
.
Taylor
DS
Fisher
MT
Turner
BJ.
2001
.
Homozygosity and heterozygosity in three populations of Rivulus marmoratus.
Environ Biol Fishes
  .
61
:
455
459
.
Taylor
DS
Turner
BJ
Davis
WP
Chapman
BB.
2008
.
A novel terrestrial fish habitat inside emergent logs
.
Am Nat
  .
171
:
263
266
.
Turko
AJ
Earley
RL
Wright
PA.
2011
.
Behaviour drives morphology: voluntary emersion patterns shape gill structure in genetically identical mangrove rivulus
.
Animal Behav
  .
82
:
39
47
.
Turko
AJ
Robertson
CE
Bianchini
K
Freeman
M
Wright
PA.
2014
.
The amphibious fish Kryptolebias marmoratus uses different strategies to maintain oxygen delivery during aquatic hypoxia and air exposure
.
J Exp Biol.
 
217
:
3988
3995
.
Turner
BJ
Davis
WP
Taylor
DS
1992
.
Abundant males in populations of a selfing hermaphrodite fish, Rivulus marmoratus, from some Belize cays
.
J Fish Biol.
 
40
:
307
310
.
Turner
BJ
Elder
JF
Jr
Laughlin
TF
Davis
WP
Taylor
DS.
1992
.
Extreme clonal diversity and divergence in populations of a selfing hermaphroditic fish
.
Proc Natl Acad Sci U S A.
 
89
:
10643
10647
.
Turner
BJ
Fisher
MT
Taylor
DS
Davis
WP
Jarrett
BL.
2006
.
Evolution of ‘maleness’ and outcrossing in a population of the self-fertilizing killifish, Kryptolebias marmoratus
.
Evol Ecol Res.
 
8
:
1475
1486
.
UniProt Consortium
.
2012
.
Reorganizing the protein space at the Universal Protein Resource (UniProt)
.
Nucleic Acids Res.
 
40
:
D71
D75
.
Valenzano Dario
R
, et al.   .
2015
.
The African turquoise killifish genome provides insights into evolution and genetic architecture of lifespan
.
Cell
 
163
:
1539
1554
.
Van der Auwera
GA
, et al.   .
2013
.
From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline
.
Curr Protoc Bioinformatics
  .
43
:
11.10.1
33
.
Wang
J
Duncan
D
Shi
Z
Zhang
B.
2013
.
WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013
.
Nucleic Acids Res.
 
41
:
W77
W83
.
Wells
MW
Turko
AJ
Wright
PA.
2015
.
Fish embryos on land: terrestrial embryo deposition lowers oxygen uptake without altering growth or survival in the amphibious fish Kryptolebias marmoratus
.
J Exper Biol.
 
218
:
3249
3256
.
Wright
PA.
2012
.
Environmental physiology of the mangrove rivulus, Kryptolebias marmoratus , a cutaneously breathing fish that survives for weeks out of water
.
Integr Comp Biol.
 
52
:
792
800
.
Yang
Z.
2007
.
PAML 4: phylogenetic analysis by maximum likelihood
.
Mol Biol Evol.
 
24
:
1586
1591
.
Yang
Z
dos Reis
M.
2011
.
Statistical properties of the branch-site test of positive selection
.
Mol Biol Evol.
 
28
:
1217
1228
.

Author notes

Data deposition : Sequence data have been deposited under NCBI BioProject PRJNA290522. This Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession LHSH00000000. The version described in this paper is version LHSH01000000.
Associate editor: Jay Storz
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( http://creativecommons.org/licenses/by-nc/4.0/ ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com