Draft Genome Sequence of Rhodosporidium toruloides CECT1137, an Oleaginous Yeast of Biotechnological Interest.

We report the sequencing of the basidiomycetous yeast Rhodosporidium toruloides CECT1137. The current assembly comprises 62 scaffolds, for a total size of ca. 20.45 Mbp and a G+C content of ca. 61.9%. The genome annotation predicts 8,206 putative protein-coding genes.

GENOME ANNOUNCEMENT

Large-scale production of biodiesel requires considerable amounts of fatty acids (FAs). The resulting demand for FAs has made oleaginous yeast species a target of choice as an alternative source of oils and fats (1). The genomic data available for such species have increased over the last few years but still remain scarce. To increase the current knowledge of oleaginous yeast species, we report the sequencing of the basidiomycetous yeast Rhodosporidium toruloides CECT1137. This strain was originally deposed in the Natural Collection of Yeast Cultures (NCYC) (United Kingdom) in 1925 by Antoine Guillermond (strain NCYC162) under the name “Levure de rose.” It was transferred in 1984 to the Spanish Type Culture Collection (CECT) (Spain), where it received the CECT1137 accession number. This strain was originally described as Rhodotorula glutinis var. glutinis based on physiological characteristics. However, our sequencing data reinstated it as a member of the species R. toruloides.

The CECT1137 genome was sequenced by Eurofins Scientific (France), using Roche 454 GS FLX Titanium, on a single-read library and an 8-kb mate-pair library. A complementary run of Illumina HiSeq 2000 was performed on a cDNA library using Illumina TruSeq RNA to improve the predictions of the gene models and their numerous introns. Several assembly runs were performed using a combination of MIRA (2) and Allpaths-LG (3). Smoothing and scaffolding were performed using PILON (version 1.7; The Broad Institute [http://www.broadinstitute.org/software/pilon/]) and SSPACE-BASIC-2.0 (4). Genes were predicted using a combination of tools, including Augustus (5), GeneMark (6), EST2Genome (7), BLASTp, and PSItblastn (8). The recently published genome of the closely related strain NP11 was integrated in the annotation pipeline as an additional support for prediction (9). The cDNA reads were mapped on the genome using Tophat2 (10). Exon-exon junctions were extracted from the alignments using a combination of in-house tools developed in BioPerl (11). Splicing events and structural annotation were manually validated using the Artemis software (12). The tRNA genes were identified using tRNAscan-SE (13).

The current draft comprises 62 scaffolds, for a total size of 20,445,260 bp and a G+C content of ca. 61.9%. Overall, 8,206 putative protein-coding genes have been identified, 212 of which harbor introns alternatively spliced within the coding sequences and/or the untranslated regions (UTR). An additional 145 genes have been annotated as dubious models or pseudogenes, with frameshifts, stops in translation, or dubious starts or stops. The genome contains 149 tRNAs and 249 miscellaneous RNAs. Whenever possible, a functional annotation was proposed, based on a combination of BLASTp (8), EMBOSS Needle (7), and InterProScan (14) against subsets of genes extracted from UniProt (15), with priority given to experimentally validated data from yeasts and fungal species. A total of 8,294 proteins were predicted (including 88 splicing variants), among which 7,080 showed at least 20% sequence similarity on 70% of an alignment with at least one gene from the Swiss-Prot/TrEMBL subsets.

Further comparison of the genome of CECT1137 against other yeast species will bring additional insights on the genomic properties of oleaginicity, providing potential targets for future biotechnological applications.

Nucleotide sequence accession numbers.

This whole-genome shotgun project has been deposited at the European Nucleotide Archive under the accession no. LK052936 to LK052997.