Due to expected climate challenges it is imperative to deepen our understanding of plant water stress tolerance mechanisms in order to bring our new crop varieties to maximal performance. Both stress response and plant metabolism are complex multi-gene traits. In order to compose a comprehensive understanding of genes regulating the metabolic response to water stress we converged three extensive “omics” methods. A large panel of tomato accessions was grown under control and reduced water conditions. Sampled leaves were submitted to global transcriptomic and high-coverage metabolomic profiling by liquid and gas chromatography-mass-spectrometry (LC/GC-MS) for polar and semi-polar and compounds. Metabolite quantification was integrated with high-density genomic SNP information provided for all accessions using genome-wide association studies (GWAS). Metabolite-gene expression correlations improved identification of target genes. This approach yielded 192 consistent metabolic quantitative trait loci (QTL) in control conditions and 186 QTL in reduced water. Of them, 38 QTL overlapped, and others were unique to each condition. One such QTL was associated with over 60 metabolites, which were significantly correlated with the expression of a gene from the little studied serine-carboxypeptidase-like family. Three other genes from this family were expressed in the leaves, each differing in their response to reduced water. In another QTL, metabolites were significantly correlated to a Purine permease family gene under water stress when gene expression was elevated. Selected QTL genes were validated by VIGS and transient OE. To conclude, the method combination produced a powerful association detection triangle allowing insight into metabolite-gene networks regulating tomato stress response.