The dataset for tomato project is still confidential; while the dataset for predictive toxicology project is available from the following two publications: (1) GE Data Waterman et al. BMC Genomics 2010, 11:9 http://www.biomedcentral.com/1471-2164/11/9 (2) Metabolite Data Anal. Chem., 2010, 82 (11), pp 4479–4485 DOI: 10.1021/ac100344m