Adaptive Monitoring of Biotechnological Processes Kinetics
Round 1
Reviewer 1 Report
Review of the article: "Adaptive Monitoring of Biotechnological Processes Kinetics".
Points for consideration:
1. In the article the Authors would like to consider a class of aerobic fed-batch processes, but to this task they use formulas (17), (18), (19) that describe continuous process (chemostat).
Moreover, as I see from Figure 3, the experiments have been carried out as fed-batch processes.
So, mathematical considerations have been conducted for continuous processes and experiments have been carried out as fed-batch processes.
2. There is question whether Bacillus subtilis should be cultivated in fed-batch process or maybe in continuous process. To answer this question, the Authors should compare Bacillus subtilis biomass productivity obtained in fed-batch and in continuous processes. It is possible that fed-batch process allows to obtain much higher Bacillus subtilis biomass productivity compared to continuous process - if so, then consideration of continuous process for cultivation of Bacillus subtilis (as it is in the article, see formulas (17), (18), (19)) does not make sense.
Author Response
Dear Editor,
We would like to thank all reviewers for their very valuable comments and suggestions. Taking into account all of them, we believe that the quality of the article has been significantly improved.
The “Track Changes” function is used during the paper revision process. Additionally, all improvements are presented in red colour in order to be easily visible to the editors and reviewers in “Simple markup” mode.
Following are the detailed answers (given in blue colour), point by point of each of the reviewers.
Reviewer 1
Points for consideration:
- In the article the Authors would like to consider a class of aerobic fed-batch processes, but to this task they use formulas (17), (18), (19) that describe continuous process (chemostat).
With respect to the remark, the Eqs. (17) - (21) describe a fed-batch process, because according to Eq. (21), dV/dt = F, the volume is not constant. It is well known that in continuous processes the volume remains constant and the dilution rate D is defined as the flow of medium per unit of time, F, over the volume V of culture in the bioreactor.
Moreover, as I see from Figure 3, the experiments have been carried out as fed-batch processes.
So, mathematical considerations have been conducted for continuous processes and experiments have been carried out as fed-batch processes.
Yes, the experiments have been carried out as fed-batch processes and the used system of differential equations (Eqs. (17) - (21)) describes the conducted fed-batch process.
- There is question whether Bacillus subtilis should be cultivated in fed-batch process or maybe in continuous process. To answer this question, the Authors should compare Bacillus subtilis biomass productivity obtained in fed-batch and in continuous processes. It is possible that fed-batch process allows to obtain much higher Bacillus subtilis biomass productivity compared to continuous process - if so, then consideration of continuous process for cultivation of Bacillus subtilis (as it is in the article, see formulas (17), (18), (19)) does not make sense.
The reviewer is absolutely right – higher biomass productivity of Bacillus subtilis is obtained during a fed-batch process. This is one of the reasons why theoretical results are proved by fed-batch experiments, carried out at Technical University – Berlin.
We have made a lot of efforts to additionally improve the English, both in grammar and style.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Modeling is a key issue in biotechnology. This research proposes the application of the model-based software sensor design to develop a new tuning procedure to reduces the number of design parameters. The research is well founded and appropriate designed.
The method, which is an extension of the method of model-based software sensor developed by Bastin and Dochain, is well described. The results are clear.
A small observation to the authors: usually, the numbering is written with letters when the number is smaller than 10 and with digits when they indicate more than ten parameters (ex. Page 6, lines 145-146).
The manuscript should be improved by adding discussion to compare the results with similar ones or others obtained with other approaches. Now, discussion is absent.
The originality is presented only in conclusion, it should be mentioned in the results, too.
Author Response
Dear Editor,
We would like to thank all reviewers for their very valuable comments and suggestions. Taking into account all of them, we believe that the quality of the article has been improved.
The “Track Changes” function is used during the paper revision process. Additionally, all improvements are presented in red colour in order to be easily visible to the editors and reviewers in “Simple markup” mode.
Following are the detailed answers (given in blue colour), point by point of each of the reviewers.
Reviewer II
Modeling is a key issue in biotechnology. This research proposes the application of the model-based software sensor design to develop a new tuning procedure to reduces the number of design parameters. The research is well founded and appropriate designed.
The method, which is an extension of the method of model-based software sensor developed by Bastin and Dochain, is well described. The results are clear.
A small observation to the authors: usually, the numbering is written with letters when the number is smaller than 10 and with digits when they indicate more than ten parameters (ex. Page 6, lines 145-146).
According the reviewer’s remark the necessary corrections have been made in the text as follows:
Page 1: row 17 (fourth and fifth order software)
Page 2: row 62 fourth order software sensor requires the determination of four parameters
row 67 (fifth order SS)
row 68 (two tuning parameters)
row 92 (is three)
Page 3: row 126 (fourth and fifth order)
Page 5: row 143 (m = 2 (fourth order system) and m = 3 (fifth order system))
row 144 (than five)
row 146 (fourth order system)
row 151 (has four different Eigenvalues)
Page 6: row 153 (a system of four)
row 154 (with four eigenvalues and six tuning parameters)
row 155 (of fifth order system)
row 158 (case of five different eigenvalues)
row 160 (of fifth order polynomial)
row 161 (presenting five algebraic)
row 162 (with five Eigenvalues and nine tuning parameters)
Page 9: row 207 (represents a fifth)
Page 9: row 228 (the following third order polynomial)
Page 11: row 239 (Experiment I and Experiment II)
Page 14: row 297 (fourth and fifth order structures)
The manuscript should be improved by adding discussion to compare the results with similar ones or others obtained with other approaches. Now, discussion is absent.
Regarding the comparison of the proposed method, estimating the yield coefficients as unknown time-varying parameters, there are no similar publications in the available literature, except for our previous studies [18, 22, 23]. They are presented in the introduction as a basis for developing the methodology proposed in the article.
The originality is presented only in conclusion, it should be mentioned in the results, too.
The following contributions have been discussed additionally:
Page 2: rows 46-53, rows 71and 74, rows 89-90
Page 4: rows 124-127
Page 14: rows 285-288
Page 14: rows 306 and 307
We have made a lot of efforts to additionally improve the English, both in grammar and style.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Dear Authors,
In continuous process:
fresh sterile substrate is dosed to bioreactor (vessel) during the process and bioreactor medium is removed during the process.
In fed-batch process:
fresh sterile substrate is dosed to bioreactor (vessel) during the process WITHOUT REMOVING bioreactor medium during the process.
There are a lot of information about continuous and fed-batch processes, for example:
https://www.onlinebiologynotes.com/cultivation-technique-of-bacteria-batch-fed-batch-and-continuous-culture-technique/
The considered by the Authors system (formulas (17), (18), (19)) presents continuous process (in particular chemostat) because fresh sterile substrate is dosed to bioreactor (vessel) during the process and bioreactor medium IS REMOVED during the process.
To validate this continuous process the Authors have used experimental data from the fed-batch process, what does not make any sense.
Moreover, I am not sure if Bacillus subtilis can be cultivated in continuous process (described by formulas (17), (18), (19)) - what I wrote in my previous review report - due to very likely decrease in biomass productivity.
Author Response
Dear Reviewer,
The mathematical description of the processes presented in the paper is based on the book of
Bastin, G.; Dochain, D. On-line estimation and adaptive control of bioreactors. Amsterdam, Oxford, New York, Tokyo, Elsevier, 1990.
As an attached file you can find a part of the book of Bastin and Dochain, explaining the basic dynamics of microbial growth in stirred tank reactors – batch (p. 5), fed-batch (p.5-6) and continuous (p. 6) reactors.
Since the dilution rate D is D = F/V,
the terms in our model system (17)-(21) “– S1.F/V”, “– X.F/V”, and “– S2.F/V” describe the decrease in the concentrations of S1, X, and S2,respectively, which is the result of dilution and not due to a REMOVED medium from the bioreactor.
Best regards,
Velislava Liubenova and co-authors
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
Dear Authors,
The article has been corrected in a proper way, among others:
- formulas (17)-(21) correctly describe fed-batch process,
- the novelties are shown in a clear way,
- the conclusions are interesting and valuable.
