Vitis vinifera has been an emblematic plant for humans since the Neolithic period. Human civilization has been shaped by its domestication as both its medicinal and nutritional values were exploited. It is now cultivated on all habitable continents, and more than 5000 varieties have been developed. A global passion for the art of wine fuels innovation and a profound desire for knowledge on this plant. The genome sequence of a homozygotic cultivar and several RNA-seq datasets on other varieties have been released paving the way to gaining further insight into its biology and tailoring improvements to varieties. However, its genome annotation remains unpolished. In this issue of Proteomics, Chapman and Bellgard (Proteomics 2017, 17, 1700197) discuss how proteogenomics can help improve genome annotation. By mining shotgun proteomics data, they defined new protein-coding genes, refined gene structures, and corrected numerous mRNA splicing events. This stimulating study shows how large international consortia could work together to improve plant and animal genome annotation on a large scale. To achieve this aim, time should be invested to generate comprehensive, high-quality experimental datasets for a wide range of well-defined lineages and exploit them with pipelines capable of handling giant datasets.