ITAG3.0 Annotation for the New Tomato Reference Genome SL3.0

Tomato (Solanum lycopersicum) is a model organism for plant biology and in particular a model system for processes such as fruit development and plant defense. Here we present an improved annotation (ITAG 3.0) for the new tomato reference genome build (SL3.0). The annotation pipeline involved updating ITAG2.40 gene models, preserving locus identifiers (solyc ids) from the previous ITAG2.40 annotation and identifying new gene models using trained ab-initio gene predictors. We successfully mapped 32,290 ITAG2.40 genes on repeat-masked SL3.0 genome. 2,379 genes out of 34,725 genes in ITAG 2.40 did not map due to improved detection methods for repeat regions. We have incorporated expression data from various sources in our pipeline including tissue and treatment specific RNAseq data, 5’ and 3’ UTR enriched RNA sequencing, RENseq for NBS-LRR genes and Pacbio Iso-seq RNA data which were kindly provided by Solanaceae community. More than 8 billion Illumina RNAseq reads were mapped to the SL3.0 genome with more than 90% mapping rate. These mapped reads were used to generate a genome-guided transcriptome. Around 2,200 genes from ITAG2.40 lack any supporting evidence such as expression or PFAM domains. The de novo prediction of genes in the maker pipeline has identified more than 2500 novel genes well-supported by expression evidence that are not present in ITAG2.40. The increased continuity of the genome in SL3.0 and availability of diverse and comprehensive expression data resulted in a significantly improved tomato annotation, including the identification of novel protein-coding genes as well as UTRs and exons not present in the previous tomato annotation. The annotation is available from SGN (https://solgenomics.net/).