Genome Sequence and Analysis of a Stress-Tolerant, Wild-Derived Strain ofSaccharomyces cerevisiaeUsed in Biofuels Research

Abstract

The genome sequences of more than 100 strains of the yeast Saccharomyces cerevisiae have been published. Unfortunately, most of these genome assemblies contain dozens to hundreds of gaps at repetitive sequences, including transposable elements, tRNAs, and subtelomeric regions, which is where novel genes generally reside. Relatively few strains have been chosen for genome sequencing based on their biofuel production potential, leaving an additional knowledge gap. Here, we describe the nearly complete genome sequence of GLBRCY22-3 (Y22-3), a strain of S. cerevisiae derived from the stress-tolerant wild strain NRRL YB-210 and subsequently engineered for xylose metabolism. After benchmarking several genome assembly approaches, we developed a pipeline to integrate Pacific Biosciences (PacBio) and Illumina sequencing data and achieved one of the highest quality genome assemblies for any S. cerevisiae strain. Specifically, the contig N50 is 693 kbp, and the sequences of most chromosomes, the mitochondrial genome, and the 2-micron plasmid are complete. Our annotation predicts 92 genes that are not present in the reference genome of the laboratory strain S288c, over 70% of which were expressed. We predicted functions for 43 of these genes, 28 of which were previously uncharacterized and unnamed. Remarkably, many of these genes are predicted to be involved in stress tolerance and carbon metabolism and are shared with a Brazilian bioethanol production strain, even though the strains differ dramatically at most genetic loci. The Y22-3 genome sequence provides an exceptionally high-quality resource for basic and applied research in bioenergy and genetics.

Nanopore sequencing enables near-complete de novo assembly of Saccharomyces cerevisiae reference strain CEN.PK113-7D

“…Fifteen of the 16 chromosomes as well as the mitochondrial genome and the 2-μm plasmid were assembled in single, mostly telomere-totelomere, contigs. This genome assembly is remarkably unfragmented, even when compared with other S. cerevisiae assemblies made with several nanopore technology flow cells, in which 18 to 105 chromosomal contigs were obtained (McIlwain et al 2016;Istace et al 2017). Despite the long-read lengths obtained by nanopore sequencing, the ribosomal DNA locus in chromosome XII could not be completely resolved.…”

“…S1). The mitochondrial genome was completely assembled, which is not always possible with nanopore sequencing (McIlwain et al 2016;Giordano et al 2017;Istace et al 2017). Even with identical DNA extraction and assembly methods, the mitochondrial genome cannot always be assembled, as illustrated by its absence in the assembly of CEN.PK113-7D Delft.…”

“…S4, Supporting Information). This is a common artefact as assembly algorithms generally have difficulties reconstructing and closing circular genomes (Venema and Tollervey 1999;McIlwain et al 2016). The coordinates of the overlaps were determined with Nucmer (Kurtz et al 2004) and manually joined resulting to a final size of 86 616 bp.…”

“…Recent developments of longread sequencing technologies have decreased the cost and increased the accuracy and output, yielding near-complete assemblies of diverse yeast strains (McIlwain et al 2016;Giordano et al 2017). For example, de novo assembly of a biofuel production S. cerevisiae strain using PacBio reads produced a genome assembly consisting of 25 chromosomal contigs scaffolded into 16 chromosomes.…”

“…Interestingly, many of these genes had functions linked to stress tolerance and carbon metabolism that are functions critical to the strains industrial application (McIlwain et al 2016). In addition, rapid technological advances in nanopore sequencing have matured as a competitive longread sequencing technology and the first yeast genomes assembled using nanopore reads are appearing (Goodwin et al 2015;McIlwain et al 2016;Giordano et al 2017;Istace et al 2017;Jansen et al 2017). For example, Istace et al sequenced 21 wild S. cerevisiae isolates and their genome assemblies ranged between 18 and 105 contigs enabling the detection of 29 translocations and four inversions relative to the chromosome structure of reference S288C.…”