Simultaneous genome sequencing of symbionts and their hosts

Abstract

Second-generation sequencing has made possible the sequencing of genomes of interest for even small research groups. However, obtaining separate clean cultures and clonal or inbred samples of metazoan hosts and their bacterial symbionts is often difficult. We present a computational pipeline for separating metazoan and bacterial DNA in silico rather than at the bench. The method relies on the generation of deep coverage of all the genomes in a mixed sample using Illumina short-read sequencing technology, and using aggregate properties of the different genomes to identify read sets belonging to each. This inexpensive and rapid approach has been used to sequence several nematode genomes and their bacterial endosymbionts in the last year in our laboratory and can also be used to visualize and identify unexpected contaminants (or possible symbionts) in genomic DNA samples. We hope that this method will enable researchers studying symbiotic systems to move from gene-centric to genome-centric approaches.

Fig. 1

Pipeline for simultaneously sequencing symbiont and host genomes. To produce high-quality assemblies of the genomes of the nematode Dirofilaria immitis and its Wolbachia wDi bacterial endosymbiont we used a preliminary assembly to sort data into bins corresponding to each genome before performing high-stringency assembly on each separately

Fig. 2

Taxon-annotated GC-coverage scatter-plot for a symbiotic system. The plot shows contigs from the Dirofilaria immitis - wDi preliminary SE assembly. Each dot represents one of 29,732 individual contigs. 10,000 contigs were randomly chosen for BLAST annotation, and 3,490 contigs with significant matches are coloured by putative taxon-of-origin. Full colour version is available online

Fig. 3

A low-complexity metagenome. Taxon-annotated GC-coverage scatter-plot (as in Fig. 2) for contigs from a preliminary SE assembly of a Caenorhabditis sp 5 DNA sample. 10,000 randomly selected contigs were BLAST-annotated by comparison to the NCBI nt database, and coloured by class-of-origin of the best match identified. Full colour version is available online