Chromosomal-Level Assembly of the Asian Seabass Genome Using Long Sequence Reads and Multi-layered Scaffolding

Abstract

We report here the ~670 Mb genome assembly of the Asian seabass (Lates calcarifer), a tropical marine teleost. We used long-read sequencing augmented by transcriptomics, optical and genetic mapping along with shared synteny from closely related fish species to derive a chromosome-level assembly with a contig N50 size over 1 Mb and scaffold N50 size over 25 Mb that span ~90% of the genome. The population structure of L. calcarifer species complex was analyzed by re-sequencing 61 individuals representing various regions across the species’ native range. SNP analyses identified high levels of genetic diversity and confirmed earlier indications of a population stratification comprising three clades with signs of admixture apparent in the South-East Asian population. The quality of the Asian seabass genome assembly far exceeds that of any other fish species, and will serve as a new standard for fish genomics.

Construction of high-resolution recombination maps in Asian seabass

“…With the exception of the fragments that had no segregated markers for alignment between LGs and chromosomes, the mismatches were verified consistently by all three family-specific genetic maps, and distributed across five chromosomes of Asian seabass [33]: Chr2, Chr8, Chr9, Chr15 and Chr20 (Fig. 3).…”

“…Integration of the integrated consensus genetic map with the contigs assembly (genome assembly version 2) and scaffold assembly (genome assembly version 3) [33], revealed that 814 out of 3807 (21.4%) contigs and 259 out of 2964 (8.7%) scaffolds and/or contigs were anchored onto the map, accounting for 560.9 Mb (88.0% of 637.5 Mb, version 2) and 579.6 Mb (90.5% of 640.2 Mb, version 3) of the total length of reference genome, respectively. Integration of genetic map with genome assembly anchored additional 275 contigs with a total length of 19.9 Mb onto all 24 LGs.…”

“…Nevertheless, the genetic architecture underlying this interesting biological phenomenon remains to be elucidated. For the past decades, a lot of genomic resources have been developed in this species to facilitate genetic improvement, e.g., hundreds of thousands of genome-wide genetic markers derived from both coding and genomic sequences [31–33], high-density genetic linkage maps [34–36], a BAC-based physical map [37] and also a chromosomal-level genome assembly [33]. Several QTL responsible for growth-related traits [34, 38, 39] and viral disease resistance [36, 40] have also been identified and used for marker-assisted selection.…”

“…Nevertheless, most of the linkage maps were constructed using microsatellites and they are of a resolution of less than 4000 markers. Thus, previous linkage maps are likely not enough to obtain confident examination of the accuracy of genome assembly and robust information on genomic heterogeneity of recombination [33, 35], and also to achieve accurate genomic prediction in the selective breeding programs of this species.…”

Genome sequence and analysis of the Japanese morning glory Ipomoea nil

“…However, a better resolution of such repeats was obtained in Oropetium thomaeum 21 genome assembly, possibly owing to the 15-kb lower end insert size selection, explaining the importance of longer read lengths in obtaining near-perfect assemblies. The potential of PacBio sequence data in long, eukaryotic genomes has been further showcased in the draft genomes of Gorilla gorilla 45 (scaffold N50 of 23.1 Mb), V. angularis 20 (scaffold N50 of 3.0 Mb), O. thomaeum 21 (contig N50 of 2.4 Mb) and Lates calcarifer 46 (scaffold N50 of 1.19 Mb). A rapid increase in PacBio sequencing for similar large-scale assemblies can be expected in the near future.…”