Stomatal closure and accumulation of wax on leaf surfaces while dehydration or desiccation tolerance has been associated with traits, such as osmotic adjustment, sugar accumulation, and maintenance of the integrity of membranes and proteins from dehydration damage. Genotypic variations in AZD6244 MEK inhibitor differential gene expression in response to drought stress are also reflected at the physiological levels. Physiological analysis with ‘Tifway’ and ‘C299’ exposed to drought stress demonstrated that ‘Tifway’ was able to maintain higher cell membrane stability and water status, as well as greater photosynthetic rate, photochemical efficiency, and antioxidant defenses. The physiological data suggested that ‘Tifway’ exhibited superior drought resistance to ‘C299’. The gene expression analysis in this study provided further insights on molecular factors associated with superior drought resistance in ‘Tifway’ bermudagrass, as manifested by the physiological traits. Previous studies have shown that proline accumulate was responsive to drought stress and serves as a protective solute to maintain cell turgor against dehydration in various plant species, oxidative protection, and function as molecular chaperone stabilizing the structure of proteins. The up-regulation of those genes associated with solute accumulation under drought stress, particularly in the drought-sensitive genotype reflected that sugar and proline accumulation was sensitive to mild or short-term drought stress in bermudagrass, but may not contribute to superior drought tolerance in this species under long-term stress. It may take part in initiating the process of leaf senescence induced by drought, which has been associated with plant survival of drought stress by reducing leaf area for transpiration to limit water loss from the plant canopy and diverting carbon partitioning. A major challenge in today’s medicine and biology is to identify the key metabolites associated with complex diseases. Because metabolites are modulated by genetic and environmental perturbations; their alterations in the concentration can reflect disturbed metabolic functions and reveal novel physiological and pathophysiological information, which can not be obtained directly from the genomics, transcriptomics, and proteomics. Metabolomics, which is a quantitative description of all endogenous metabolites found in cells and body fluid, aims at characterization of the metabolome under different conditions. Metabolomics can not only help us illustrate the underlying molecular disease-causing mechanisms but also gain broad recognition in discovery of metabolic signatures for disease diagnosis. However, these high-throughput techniques have several limitations. For example, it is difficult to determine quantitative information from peak integration due to the different ionization ability of various metabolites and the sensitivity of these techniques is not satisfactory, which can lead to false positive metabolomics results. Therefore, it is necessary to develop a computational method to prioritize the candidate disease metabolites from metabolomics profiles. The development and completeness of some high quality metabolic network databases have led to availability of computational method for prioritization of metabolites.