Question

o Flipped class: Michaelis-Menten vs. Lineweaver-Burk 10 Essentially, we'll duplicate the error estimates from (A) a nonlinear fit and (B) a nonlinear function transformed into linear form in Matlab. 1) Use Matlab to generate synthetic data obeying the Michaelis-Menten equation i.e. find dP/dt for [S] = 1:20. Add noise to the data (rand or randn or normmd). Use Vmax-1, km-5 2) Plot the data points (dP/dt vs [S]) that you've obtained. Fit the data (model1:- fitnlm(x,y.modelname,jnitialguesses). Output the estimated Vmax and km with standard errors from the fit (model1.Coefficients.Estimate). Plot the result on top of the data 3) Transform the same data into Lineweaver-Burk form: dt max IS]Vmax Fit these transformed data using either nlinfit (model2-fitnlm(x.y,modelname.initialguesses) or polyfit. 4) Plot the transformed data points (aPidt vs. V¡s) and the fit. Again output the estimated Vmax and km with standard errors from the fit (mode/2. Coefficients Estimate) Are the values better using a nonlinear fit to raw Michaelis-Menten data, or a linear fit using data transformed using Lineweaver-Burk'? Note to self: LeastSquaresFitWithErrors.nb, Matlabcode.m

Answer 1

Non - linear function transmitted in to linear form f . f Michaelis - Menten % S + E rightarrow % ES rightarrow P + E Rate k cal % ES rightarrow S + E Rate k Backward % Rate rightarrow V_max ks/(km = s) % where % V_max = k - cat times E - total % k m = (k - backward + k cat)/k forward Mat lab code: km = 5: s = 0: 1: 20: v_max = 1: nm = @ (s, v_max, km) V_max times s ((k m + s): mm- 1 = @ (s) mm (5, 1, 20): figure (1) f plot (mm - 1, [0, 3])