|
|
|||
|
|
|
|
|
|
||||
|
Company History Key People News Newsletter Customer Profiles Contact Us Employment |
Category: Biochemistry Biochemists used statistical analysis software to learn how greenhouse gases contribute to global change Creating a Predictive Mathematical Model One of the big questions about greenhouse gases and their contribution to global climate change is the rate at which organisms in the ocean consume oxygen and produce carbon dioxide. Biochemists Ted Packard, Sylvie Roy, and their colleagues at the Maurice Lamontagne Institute (MLI) in Mont-Joli, Quebec have been modeling several of the biochemical processes that take place during the respiration of plankton and bacteria. (This research began during Ted's seven-year stint at the institute, and he continues it on a free-lance basis at the Institut Cienciès del Mar in Barcelona.) The researchers set out using data at hand to crystallize current knowledge into a predictive mathematical model of respiration throughout all growth phases of a bacterial culture. This model is based on the idea that CO2 production is largely controlled by one enzyme, isocitrate dehydrogenase, and by the availability of this enzyme's reactants. To express this control mathematically, it is necessary to determine values for the affinity of this enzyme for its reactants as well as values for the concentrations of these reactants in the bacteria cell. The original model calculated the reactants as functions of time from the carbon in the bacteria's food source and the bacterial biomass. In addition, it estimated the affinity of the enzyme for its reactants from measurements made on other organisms. The model worked very well in predicting the respiratory CO2 production in all phases of a bacterial culture. But if the estimations of the affinities and other key variables could be replaced by measurements the model could be put to a more rigorous test. Author Roy's doctoral thesis has done exactly this. She measured the enzyme affinities and the concentrations of the controlling reactants in bacterial cells. With this data in hand, she introduced her measurements into the model and ran it again. While in doing so she changed a number of things, such as the mathematical form of a few algorithms, her results supported the group's original findings: respiratory CO2 production at the physiological level can be calculated from biochemical measurements of enzyme activity, the availability of the enzyme's reactants, and the affinities of the enzyme for these reactants. Packard, Roy, and their colleagues used information from laboratory experiments to measure CO2 production in bacterial cultures grown in the laboratory. Based on biochemical data from the interworkings of the bacteria's cytoplasm they developed a model for respiration of the intact, whole bacterial cell; they then tested and refined their model using the visual statistical analysis package Data Desk. Optimization was achieved by successive iterations in Data Desk using the sliders. [Sliders are interactive tools that allow the parameters of a function to be changed, and the effects observed, in real-time. In this case, they are connected to the program's nonlinear modeling command.] Then the model was run on the second set of data. The fit to the second set of data was not as good as the first, but showed that the model was on the right track. The researchers were able to enter a first-draft function for the model, compare the output of the function with experimental observations, then alter it with the sliders to fit it more closely to the data. "By running these analyses over and over," Packard said, "you gradually refine the coefficients in the equations until you have a reasonable mathematical description of this physiological process." Sylvie O. Roy is engaged in postdoctoral research at the Maurice Lamontagne Institute in Mont-Joli, Quebec. Ted T. Packard teaches biochemistry and carries out research at the Insitute Cienciès in Barcelona. |
Name: Sylvie Roy and Ted Packard Affiliation: Maurice Lamontagne Institute Location: Mont-Joli, Quebec
|
|||
|
|||||
|
|
|||||
