Spirulina-in Silico-Mutations and Their Comparative Analyses in the Metabolomics Scale by Using Proteome-Based Flux Balance Analysis

Round 1

Reviewer 1 Report

Major

1) Correlation of GSMM data to experimental data is in a large supplemental table. The authors should do a correlation analysis and also summarize the data tables. As presented, it is impossible to effectively evaluate how their model is performing as they have not adequately analyzed their model performance.

2) The metabolic flux changes in Fig 2 are difficult to understand because the annotation of the pathways in the figure is not well done.

Figure 3 is not consistent with Figure 2. Figure 3 is much easier to read, though the pathways are not well labelled to understand what is going on easily.

3) I fundamentally disagree with using protein concentrations as a proxy for metabolic flux. A good reference for how to correlate proteome data with metabolic flux comes from Chlamydomonas (e.g. Weinkpp et al. 2010, DOI: 10.1039/B920913A), where they use HPC measurements. These reactions are not governed by protein concentration since they are the catalyst, they are instead governed by metabolic concentrations and the enzyme kinetics.

The model that the authors present does not explicitly correlate to the enzyme kinetic parameters that would be critical for estimating metabolic flux, to the utility of their GSSM is of limited value. Perhaps the authors did account for this, but without a code evaluation of their model it appears that their GSSM model is quite limited compared to prior efforts on this front.

4) The analysis of the in silico mutants is interesting and consistent with prior results, but I am still not convinced that this paper is significantly advancing our understanding of algal biomass accumulation, nor providing a systems understanding of it to make significant predictions.

Minor

1. Introduction, use common gene names first when possible, e.g. histidine kinase (Hik28, SPLC….)

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this work, Lertampaiporn et al constructed a proteomics-based genome scale model of A. platensis. The work is well-executed overall and interesting for the field. There are a few issues with the articulation of the work that I mention in the following points:

• in line 24, the words "in silico" should be added before "metabolic engineering effort
• Figure 2 is a bit uninformative. The authors want to give an overview of metabolic variations in the different conditions, however the lack of annotation of at least the major pathways or major reactions makes this information impossible to be communicated (unless the reader knows by heart the kegg metabolic map). Also, the green lines are impossible to discern
• Throughout the manuscript, it is not clear whether the authors generated the proteomics datasets as part of this work or used already published datasets. Even in the case of previous dataset use, the authors should include some information on the content and quality of the protein datasets
• The authors use a protein-based model, which they rightly argue that has some advantages (and disadvantages) over other model generation methods. In this case, did the authors verified the flux values they observe by comparing to experimental data (e.g. some C13 data or following the kinetics of some key metabolites)?
• In conditions of reduced growth, we expect of course reduced fluxes in most reactions. If the authors normalize the metabolic reactions to the different growth (biomass producing) rates does this picture stand?
• in the in silico mutagenesis section, what do the authors actually mean by mutagenesis? In my understanding mutagenesis means the introduction of mutations in the protein sequence, which is not what the authors do. By running back to the methods section, the authors describe their in silico knockout methodology; they should use the same term in the results section.
• As a more minor comment, the conclusion section could be slightly expanded to give more clearly the insights their modeling studies give. What difference did they see in the different conditions, also between the 25/40 degree data? What new info did they get about the TCA cycle function and what hypotheses they can draw?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All of my concerns have been addressed by the authors.