Long-read only assembly of Drechmeria coniospora genomes reveals widespread chromosome plasticity and illustrates the limitations of current nanopore methods

Abstract

Background Long-read sequencing is increasingly being used to determine eukaryotic genomes. We used nanopore technology to generate chromosome-level assemblies for 3 different strains of Drechmeria coniospora, a nematophagous fungus used extensively in the study of innate immunity in Caenorhabditis elegans. Results One natural geographical isolate demonstrated high stability over decades, whereas a second isolate not only had a profoundly altered genome structure but exhibited extensive instability. We conducted an in-depth analysis of sequence errors within the 3 genomes and established that even with state-of-the-art tools, nanopore methods alone are insufficient to generate eukaryotic genome sequences of sufficient accuracy to merit inclusion in public databases. Conclusions Although nanopore long-read sequencing is not accurate enough to produce publishable eukaryotic genomes, in our case, it has revealed new information about genome plasticity in D. coniospora and provided a backbone that will permit future detailed study to characterize gene evolution in this important model fungal pathogen.

Chromosomal rearrangements with stable repertoires of genes and transposable elements in an invasive forest-pathogenic fungus

“…Associated with the npScarf scaffolding tool, it was the most satisfactory method for assembling this genome. In addition, our study confirmed the importance of performing several steps of manual curation, and identifying possible chimeric scaffolds (Courtine et al 2020). However, several repeated regions along the genome could still be imperfectly assembled (estimated to 0.5 Mb, little more than 1% of the genome).…”

Antagonistic fungal enterotoxins intersect at multiple levels with host innate immune defences

“…Dozens of the D. coniospora proteins predicted to be secreted are lineage-specific [2] and presumable result from co-evolution with nematode hosts. They are of great interest for the understanding of evolutionary dynamics, for which D. coniospora is potentially a powerful model [11], but represent a major challenge due to the lack of any prior knowledge. Therefore, here, we chose to focus on enterotoxin genes that are expected to play a direct role in fungal virulence.…”

“…A Circos plot [108] showing the positions of the predicted enterotoxin genes in the genomes of 2 D. coniospora strains. The full isolation history of the strains Swe2 (left hand side, derived from ATCC 96282 [2]) and Dan2 (right hand side, ARSEF 6962 [109]) are given elsewhere [11]. Orthologous gene pairs, with their corresponding Genbank protein identifiers, are joined by lines, coloured on the basis of the position on Swe2 chromosomes.…”

Antagonistic fungal enterotoxins intersect at multiple levels with host innate immune defences

“…Dozens of the D. coniospora proteins predicted to be secreted are lineage-specific and presumable result from co-evolution with nematode hosts. They are of great interest for the understanding of evolutionary dynamics, for which D. coniospora is a potentially powerful model (Courtine et al, 2020), but represent a major challenge due to the lack of any prior knowledge. Therefore, here, we chose to focus on proteins that on the basis of conserved domains were expected to play a direct role in fungal virulence.…”

“…D. coniospora (Swe3, derived from ATCC 96282 (Courtine et al, 2020)) spores were amplified by infecting worms every one or two weeks in the lab. The method to grow spores is described in detail in (Powell and Ausubel, 2008).…”