Thede novoreference genome and transcriptome assemblies of the wild tomato speciesSolanum chilense

BackgroundWild tomato species, likeSolanum chilense, are important germplasm resources for enhanced biotic and abiotic stress resistance in tomato breeding. In addition,S. chilenseserves as a model system to study adaptation of plants to drought and to investigate the evolution of seed banks. However to date, the absence of a well annotated reference genome in this compulsory outcrossing, very diverse species limits in-depth studies on the genes involved.FindingsWe generated ∼134 Gb of DNA and 157 Gb of RNA sequence data ofS chilense, which yielded a draft genome with an estimated length of 914 Mb in total encoding 25,885 high-confidence (hc) predicted gene models, which show homology to known protein-coding genes of other tomato species. Approximately 71% (18,290) of the hc gene models are additionally supported by RNA-seq data derived from leaf tissue samples. A benchmarking with Universal Single-Copy Orthologs (BUSCO) analysis of predicted gene models retrieved 93.3% BUSCO genes, which is in the current range of high-quality genomes for non-inbred plants. To further verify the genome annotation completeness and accuracy, we manually inspected the NLR resistance gene family and assessed its assembly quality. We revealed the existence of unique gene families of NLRs toS. chilense. Comparative genomics analyses ofS. chilense, cultivated tomatoS. lycopersicumand its wild relativeS. pennelliirevealed similar levels of highly syntenic gene clusters between the three species.ConclusionsWe generated the first genome and transcriptome sequence assembly for the wild tomato speciesSolanum chilenseand demonstrated its value in comparative genomics analyses. We make these genomes available for the scientific community as an important resource for studies on adaptation to biotic and abiotic stress inSolanaceae, on evolution of self-incompatibility, and for tomato breeding.