Simulating DBR Biotransformation

From ISMOC
Revision as of 15:36, 15 August 2016 by Aliah.hawari (talk | contribs) (Average Product Ratio)
Jump to: navigation, search
Figure 1: Overview of simulating DBR biotransformation from Toogood, et. al, 2015.

SUMMARY This DBR simulation is of the biotransformation experiment as described in Toogood, et. al. (2015) [1]. The experiment observed a percentage yield of 40% and 33% of menthone and isomenthone respectively after 24 hours. An overview of the simulation including the concentration of enzyme, substrate and cofactor is shown in Figure 1. Ensemble modelling of this reaction was carried out for 200 samples where the kinetic parameters were sampled randomly from 7 probability distributions. To describe the production variability of menthone to isomenthone, a variability factor (X1 and X2) were included in the reaction equation. These variabililty factors were calculated from the average product ratio sampled from a probability distribution.





















Average Product Ratio

Figure 2: Probability distribution of average product ratio values

Average product ratio probability distribution

Yield ratio of menthone:isomenthone were reported in Toogood, et. al.(2015) [1] (40:33), Solodar (1976) [2] (70:30) and Ringer, et. al. (2003) [3] (55:45). The global average product ratio of these values were taken as the mode and a CI factor of 10 was used to generate probability distribution. This probability distribution is used to sample random average product ratio during ensemble modelling.

Figure 2 shows the probability distribution of 1000 sample values for the average product ratio of DBR. The mode was determined to be 1.59 and the sample values are within the range of 0 to 100. Average product ratio between 0.16 to 15.9 falls within the 95% of confidence interval. The average product ratio observed from Toogood, et. al. (2015) [1] is 1.2 for yield of 40% : 33% of menthone:isomenthone respectively.

Average product ratio samples

Figure 3: Kernel density plot of average product ratio sampled

An ensemble modelling of 200 DBR samples was carried out, which means 200 of DBR average product ratio values were sampled from the probability distribution (Figure 1). Figure 2 shows the distribution of the 200 average product ratio values sampled where the median of these values was used as the mode (3.6). The peak is at about 100 (frequency axis), which means that the majority of the average product ratio sampled are less than 5 and these values (in the peak bin) account up to 55% of the total sample.

As the observed average product ratio from Toogood, et. al. (2015) is 1.2, it falls within this peak kernel.

References