Accurate, Multi-Kb Illumina Truseq Synthetic Reads Resolve Complex Microbial Populations and Detect Rare Microorganisms

Genomic analysis of complex microbial communities in environments such as soils and sediments remains challenging, despite increasing throughput of short read DNA sequencing technologies. Long read sequencing technologies have the potential to assist in metagenomics efforts through improved genome assembly and the analysis of sufficiently long unassembled reads. In this talk I will discuss our experience of studying complex microbial communities from terrestrial sediments using short Illumina HiSeq reads in tandem with long Illumina TruSeq (Moleculo) Synthetic Long-Reads. Overall, ~200 Gbp short and ~1.5 Gbp long read data were generated for three sediment samples at median read lengths of 150 and ~8,000 bps, respectively. Relatively small amounts of long read data assembly were achieved due to very low coverage for even the most abundant genomes. A significant portion of the long reads (40-65%) represented genomes that were too rare to be assembled using the short read data, thus providing a glance into the “long tail” of rare organisms in our samples.  The most abundant operational taxonomic units (OTUs) in all samples were represented by heterogeneous mixtures of strains and species with very little representation in the short read assemblies. Reconstruction of most of the genome architecture for these OTUs in two of the samples was made possible through overlap-based clustering of the long reads and synteny analysis of genes on those reads. Overall our results show that pairing long and short read technologies provides a combination of coverage depth and read length that enables accurate analysis of microbial communities.