Employing Active Learning in Medium Optimization for Selective Bacterial Growth

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The efficiency and practicality of active learning in medium optimization for selective culture have been demonstrated here. It has successfully offered some insights into the contribution of the chemical components to specific bacterial growth. Correct bacterial cultures are essential for isolating and functionalizing individual bacteria in microbial communities. However unknown mechanisms between bacterial growth and medium components remain a challenge. In this study, machine learning (ML) has been combined with active learning to finetune the medium components for the selective culture of two divergent bacteria, i.e., Lactobacillus plantarum and Escherichia coli. Python is used for ML. The methods have been elaborated briefly. The growth parameters here have been designed as the exponential growth rate and maximal growth yield, which were calculated according to the growth curves. The bacterial culture-based biological assays have also been provided with authentic information and valid analyses.

Results are presented professionally and the connection between biology and ML models has successfully been made. The introduction is rigorous. The concluding remarks in the discussion are clear and sound.

I have no reason against giving a green signal. Congratulations!

 

 

 

Author Response

Thank you. We are happy that you do understand our study very well.

Reviewer 2 Report

Comments and Suggestions for Authors

Authors have presented the Employing active learning in medium optimization for selective bacterial growth. The concept of MS is good. However, the manuscript can be enhanced through strengthening the experimental segment, and in particular, the conclusion.

The specific comments, which could help to improve the manuscript are:

1.      Manuscript should be revised for grammatical & punctuation errors.

2.      In introduction, add the reason behind the selection of both the bacterial strains.

3.      Figures 1-8 are missing.

4.      It looks that study lacks any laboratory experiment.

 

5.       Authors did not conclude their findings and future directions. Add conclusion section.

Comments on the Quality of English Language

Moderate editing of English language is required.

Author Response

We are sorry that our study seemed to be out of your research scope. 

Reviewer 3 Report

Comments and Suggestions for Authors

The study utilizes machine learning and active learning to optimize the medium for selective bacterial culture, demonstrating improved specificity in Lactobacillus plantarum and Escherichia coli growth while uncovering key medium components influencing growth decisions. The referee suggests article minor revision.

After the manuscript resubmission and the inclusion of figures, the document now exhibits enhanced clarity. However, several queries have been identified:

1. Could you provide more detailed information about the criteria selected for experimental verification?

2. It is recommended to articulate the research objective in a separate paragraph at the conclusion of the introduction section.

3. Clarification is needed on the selection process for medium components. Additionally, details on the methods employed to estimate their contents would be valuable.

4. Given the considerable concentration of the media, which has the potential to elevate the ionic strength (µ, M) of the solution, it would be beneficial to elaborate on how KH₂PO4 resulted in improved outcomes.

5. For the sake of currency, please incorporate references from novel literature published after July 2023.

Your attention to these points will contribute to the overall refinement of the manuscript.

Author Response

Response to Reviewer

Thank you for your careful reading and valuable comments, which helped us improve the manuscript to a great extent. The changes made are highlighted in the manuscript.

 

1. Could you provide more detailed information about the criteria selected for experimental verification?

Response: After each round of ML prediction, the predicted results were sorted from high to low according to the target score. The medium combinations (compositions) corresponding to the top 10-20 scores were experimentally verified, as described in the Materials and Methods (lines 364-366).

 

2. It is recommended to articulate the research objective in a separate paragraph at the conclusion of the introduction section.

Response: Thank you for your suggestion. The research objectives were added, and the corresponding paragraph was revised in the Introduction (lines 72-77).

 

3. Clarification is needed on the selection process for medium components. Additionally, details on the methods employed to estimate their contents would be valuable.

Response: Thank you for your suggestion. Selecting the MRS medium (11 components) was to benefit from the machine learning techniques. A brief explanation was added to the Results (lines 99-103).

 

4. Given the considerable concentration of the media, which has the potential to elevate the ionic strength (µ, M) of the solution, it would be beneficial to elaborate on how KH2PO4 resulted in improved outcomes.

Response: It’s intriguing why KH2PO4 played such a crucial role in selective culture. At the moment, we don’t know the reason why. As the present study reported a novel methodology but not the molecular mechanism, clarifying the changes in ionic strength in the solution and the specific metabolic processes of bacteria are out of the scope of this research. Instead, the possible involvement of K2HPO4 and other components participating in metabolism was discussed (lines 214-220).

 

5. For the sake of currency, please incorporate references from novel literature published after July 2023.

Response: Thank you for your suggestion. Five new citations (refs. 12, 18, 21, 22, 25), which were reported after July 2023, were added.