Benchmarking Oxford Nanopore read assemblers for high-quality molluscan genomes

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

14 Scopus Citations
View graph of relations


  • Jin Sun
  • Runsheng Li
  • Chong Chen
  • Julia D. Sigwart
  • Kevin M. Kocot


Original languageEnglish
Journal / PublicationPhilosophical Transactions of the Royal Society B: Biological Sciences
Issue number1825
Online published5 Apr 2021
Publication statusPublished - 24 May 2021


Choosing the optimum assembly approach is essential to achieving a high-quality genome assembly suitable for comparative and evolutionary genomic investigations. Significant recent progress in long-read sequencing technologies such as PacBio and Oxford Nanopore Technologies (ONT) has also brought about a large variety of assemblers. Although these have been extensively tested on model species such as Homo sapiens and Drosophila melanogaster, such benchmarking has not been done in Mollusca, which lacks widely adopted model species. Molluscan genomes are notoriously rich in repeats and are often highly heterozygous, making their assembly challenging. Here, we benchmarked 10 assemblers based on ONT raw reads from two published molluscan genomes of differing properties, the gastropod Chrysomallon squamiferum (356.6 Mb, 1.59% heterozygosity) and the bivalve Mytilus coruscus (1593 Mb, 1.94% heterozygosity). By optimizing the assembly pipeline, we greatly improved both genomes from previously published versions. Our results suggested that 40-50X of ONT reads are sufficient for high-quality genomes, with Flye being the recommended assembler for compact and less heterozygous genomes exemplified by C. squamiferum, while NextDenovo excelled for more repetitive and heterozygous molluscan genomes exemplified by M. coruscus. A phylogenomic analysis using the two updated genomes with 32 other published high-quality lophotrochozoan genomes resulted in maximum support across all nodes, and we show that improved genome quality also leads to more complete matrices for phylogenomic inferences. Our benchmarking will ensure efficiency in future assemblies for molluscs and perhaps also for other marine phyla with few genomes available. This article is part of the Theo Murphy meeting issue 'Molluscan genomics: broad insights and future directions for a neglected phylum'.

Research Area(s)

  • assembly, molluscan genomes, Mytilus, Oxford nanopore technology, phylogeny, Scaly-foot Snail