The Pycnogonida, or sea spiders, are essential arthropod taxa for understanding the early evolutionary history of arthropods and relationships between primary clades of this phylum. Studies based on both molecular and morphological data suggest that sea spiders have archaic origins as old as the early Cambrian. They are also considered an early-branching lineage with a deep split origin, which is implied by their incomplete early fossil records and relatively recent diversification of extant members (Dunlop and Selden, 2009; Rota-Stabelli et al., 2013; Sabroux et al., 2019; Lozano-Fernandez et al., 2020). However, their peculiar morphologies such as extremely reduced trunk and 8–12 walking legs containing part of their digestive and reproductive organs are highly derived and very unusual among arthropods (Sabroux et al., 2019). With a combination of the molecular phylogenetic studies using the limited numbers of marker genes, their peculiar adult and developmental morphologies had led the Cormogonida hypothesis (Zrzavý et al., 1998; Giribet et al., 2001; Dunlop and Arango, 2005; Machner and Scholtz, 2010), which places pycnogonids as a sister taxon to all other arthropods, as opposed to the monophyletic Chelicerata hypothesis, which recognizes their sister taxon relationship with the Euchelicerata. Unlike these molecular phylogenetics studies based on a few markers, numerous phylogenomic studies based on thoroughly sampled expressed sequences tags repeatedly designated sea spiders as basalmost chelicerates (Meusemann et al., 2010; Regier et al., 2010; Rehm et al., 2011). The most recent phylogenomic researches incorporating significantly increased taxa and tested loci based on the whole-genome or transcriptome data (Sharma et al., 2014; Ballesteros and Sharma, 2019; Lozano-Fernandez et al., 2019) have brought consensus on pycnogonids firmly nested in the Chelicerata.

These recent phylogenomic studies provided newly sequenced chelicerate genomes or transcriptomes, in addition to invaluable insights on the evolution of Euchelicerata (Sharma et al., 2014; Ballesteros and Sharma, 2019; Lozano-Fernandez et al., 2019). However, studies using such an approach on the Pycnogonida are currently very limited and fail to incorporate a de novo assembled pycnogonid genome (Dietz et al., 2019; Ballesteros et al., 2020). This lack of pycnogonid genome assembly is also a challenging obstacle for studies on the evolution of their distinctive morphology that requires high-quality genome assemblies with reliable annotations (Garb et al., 2018). To the best of our knowledge, no pycnogonid genome has been sequenced to date, and there are only three cases of unpublished flow cytometry reports on pycnogonid species so far (Animal Genome Size Database, https://genomesize.com). We aimed to provide an annotated pycnogonid genome sequence using the PacBio long-read de novo genome sequencing to provide useful information for understanding the evolution of unique pycnogonid morphological and developmental characteristics. Here, we report a draft of the first high quality annotated genome of a common sea spider species, Nymphon striatum, in Korean waters. This is the first de novo sequenced and assembled genome that represents the Class Pycnogonida.

The genome size of N. striatum was estimated at ~607 Mb, while the genome survey result showed relatively high (~1.9%) heterozygosity ratios and prominent heterozygosity peaks clearly separated from the homozygosity peaks (Supplementary Figure 1). The almost equal distributions of non-duplicated (1X) and duplicated (2X) K–mers at the homozygosity peak further supported the highly heterozygous nature of its genome (Supplementary Figure 3) that we speculated the pooling of N. striatum specimens was likely its main cause. The initial contig–level genomic assembly of N. striatum was 1.26 Gb long, of which the total base length exceeded twice those of the genome survey results (Table 1A). Genome assembly assessment using Benchmarking Universal Single-Copy Orthologs (BUSCO) (Simão et al., 2015) also showed that 53.47% of BUSCO genes from the initial assembly were duplicated, which further indicates the presence of numerous redundant haplotypic contigs (Table 1B). Therefore, these haplotypic contigs were reduced by applying Purge Haplotigs (Roach et al., 2018) and Redundans (Pryszcz and Gabaldoón, 2016). The Purge Haplotigs-curated contig–level draft genome demonstrated prominent improvements in its total length, which was reduced to 732.9 Mb, and contig N50 almost doubled to 360.90 Kb, with its number of contigs being reduced to 2,946 (Table 1A). Using arthropod database, the BUSCO assessment of the revised contig–level assembly resulted in a substantial improvement: the percentage of complete, but duplicated BUSCO genes was reduced to 16.79% (Table 1B and Supplementary Figure 2). The K–mer distribution analyses using the K–mer analysis toolkit (Mapleson et al., 2017), further found that the ratios of heterozygous haplotigs (indicated by 2X, or duplicated K–mers) were greatly reduced at the homozygosity peak after running Purge Haplotigs (Supplementary Figure 3). After conducting scaffolding and gap closing, the final version of the scaffold–level draft genome was composed of 1,638 scaffolds with scaffold N50 increased to 701.80 Kb and a minute fraction (0.04%) of N bases (Table 1A). We found that 31.9% of the genomic scaffolds were longer than 500 Kb, and among them, 159 scaffolds were over 1,000 Kb. In particular, only seven scaffolds were shorter than 10 Kb, which indicated that our genome was highly contiguous (Figure 1A).

An ab initio repeat element prediction resulted in a total of 7.14% of the genomic sequences annotated as repetitive sequences (Table 1C). Among these repeat elements, short interspersed nuclear elements (SINEs) accounted for 28.20% of the total length of the annotated repeat sequences, and they were recorded as the most enriched repeats from the N. striatum genome. The SINEs were followed by simple repeats (24.09%), small RNAs (21.35%), and non-SINE interspersed elements (20.82%), while satellites (0.36%) occurred the least among the categorized repeat elements (Figure 1B). The final N. striatum genome consisted of 28,539 genes which spanned 56.01% of the total genomic length. Additionally, 14,247 transfer RNA and 308 ribosomal RNA genes were annotated (Table 1D). The summed repetitive element contents of 7.14% and gene span of 51.06% were significantly different when they were compared to those of other chelicerate genomes used in the basic comparative analyses of this current study. The ortholog analysis conducted using six ecdysozoan genomes (Supplementary Table 3) resulted in 7,597 orthologous clusters and 3,888 singletons for N. striatum, with 4,493 orthologous clusters shared within four chelicerate species (Figure 1C). Phylogenetic tree reconstruction strongly supported the earliest-branching position of N. striatum nested in the monophyletic Chelicerata (Figure 1D) with maximum bootstrap support values (100%) and posterior probabilities (1.0). Nevertheless, more terrestrial arachnid taxa are required to be sampled in further analysis to provide a conclusion about a contentious phylogenetic position of horseshoe crabs deeply nested in the Arachnida that the recent phylogenomic studies on the Euchelicerata have debated (Sharma et al., 2014; Ballesteros and Sharma, 2019; Lozano-Fernandez et al., 2019). To sum up, we suggest that the draft N. striatum genome that we assembled is nearly complete and provides the first reference genome representing pycnogonids.

The datasets generated for this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI under accession numbers PRJNA595812 and SAMN13567730.

WK and SR conceived the study. J-HJ and HK performed the bioinformatics analyses and wrote the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.