Creating/Editing reactions

When you select "Create New Reaction" or click on a reaction name in the table an editor window opens.

WeDesign allows modifying the name of the reaction, the stoichiometry coefficients of each substrate and product, adding or deleting of compounds, marking the reaction as reversible or including its flux constraint. Simply start typing into the substrates/products text field and CyanoFactory will suggest metabolites based on your input. To change the coefficients click on the metabolite.

Creating of new metabolites is not possible in this view. You have to create them with " "#Create Metabolite" beforehand.

For creating multiple reactions at once we recommend using the "Batch create" functionality described below.

Batch creating of reactions

To reach this function click on the arrow next to "Create reaction" and then "Bulk create reactions".

The batch create option allows you to paste reactions in BioOpt format and it will automatically create all reactions and new metabolites for you.

The BioOpt format is:

reaction name : 2 substrate_1 + substrate_2 -> product_1 + 2 product_2

Each line starts with the reaction name followed by a ":". On the left side of the arrow "->" you can put substrates delimited by +. In front of the substrates is the quantity which can be omitted for a quantity of "1". Right to the arrow are the products which follow the same rules as the substrates. To make a reaction reversible use "<->" as the arrow.

A summary below the text box will show how WeDesign parsed the string. If the parsing looks correct click "Create reactions" to close the window and return to the editor.

Filter

The reaction list can be filtered using three different predicates:

• Active/Inactive: Only show reactions that are enabled/disabled
• Constraint/Unconstrained: Only show reactions whose fluxes are constraint or unconstrained.
• Reversible/Irreversible: Only show reactions that are reversible (<->) or irreversible (->)

Search function

The search function does a full text search over the whole reaction (including metabolites) and allows filtering of entries which contain the provided text.

For more advanced searching and filtering enable "Search with RegExp" which will find entries based on whether a regular expression matches. Please consult a guide about regular expressions for usage.

Filter

The metabolite list can be filtered based on "Is internal" and "Is external" and will only show internal or external metabolites respectively.

Search function

The search function does a full text search over the whole metabolite list and the reaction names in the consumed/produced columns and allows filtering of entries which contain the provided text.

For more advanced searching and filtering enable "Search with RegExp" which will find entries based on whether a regular expression matches. Please consult a guide about regular expressions for usage.

Creating/Editing metabolites

When you select "Create New Metabolite" or click on a metabolite name in the table an editor window opens.

WeDesign allows modifying the name of the metabolite and marking it as external.

Deleting metabolites

Deleting of a metabolite is not possible when a metabolite is in use by a reaction. When metabolites are unused you can click "Delete unused metabolites" to remove all of them at once.

Flux Balance Analysis (FBA)

Flux Balance Analysis will Maximize or Minimize the objective function using linear programming and a graph (reactions as edges, metabolites as nodes) showing the calculated fluxes of all reactions will be displayed.

Due to the slowness of the graph algorithm the graph will only show the 30 reactions with the highest flux which are connected with the objective. To change this behaviour enable the option "Display selected reactions" and manually pick the reactions you are interested in.

Metabolic Burden Analysis (MBA)

Sensitivity Analysis (SA)

Introduction

This model pretends to be the first hands-on experience to understand and simulate genome-scale metabolic models. The model is fictitious and simple in order to make the whole process easier. It consists of 7 reactions and 8 metabolites, among them being three external ones, which are A(ext), D(ext) and E(ext).

For selecting the toy model, we:

• As a guest: Select "Toy Model" from the model list
• As a registred user: Select "Create New Model -> From Template" and choose "Toy Model".

Optimisation of one metabolite production

We can now simulate the set of fluxes that maximise the output of reaction 5 by increating the production of metabolite D. Please remember that in FBA simulations, there is always the need of a transport reaction that makes an intracellular compound to become extracellular, if you want to optimise its flux.

For maximising that reaction, we first constrain reaction 1, the uptake reaction for metabolite A, within the flux range [0, 1] and simulate optimising reaction 5, setting this as Main Objective.

The uptake of metabolite E is constrained to [0, 3] units. As we can see in the results image, this uptake flux directly maximises the output of D, so the algorithm takes the maximum flux for E(ext) to maximise D(ext). On the other side A(ext) is also contributing to maximise D(ext), so the maximum allowed metabolite A(ext) uptake flux is used.

We will repeat the simulation at this moment but changing the maximum flux value for A(ext) uptake from 1 to 10 and simulate again:

The results show that now the main flux contribution to D(ext) comes from the metabolic pathway involving A(ext).

TASK: Try now to constrain now both A(ext) and E(ext) uptake reactions to 1 unit and analyse the results. If we obviate energetic and kinetic properties, which metabolite is more suitable for producing D in the cell?

Optimisation of the production of two metabolites

The next exercise consists of modifying the metabolic network by adding the following reactions:

reac8: E + B -> 2 F reac9: F -> F ext

Use the bulk add functionality to add them quickly.

We will use now Metabolic Burden Analysis: We will maximise now the flux of F(ext), setting it as Design Objective, by fixing the flux of D(ext) to different values: 0, 5, 10, 15 and 20 units (that is D(ext) = [0,0], [5,5]...) and simulate for each of the given D(ext) fluxes. For this, you will need to set first the constraints for both uptake reactions related to A(ext) and E(ext) to 10 units. After having done these 5 simulations and plotting the results of the fluxes for D(ext) and F(ext) together, the following graph can be obtained:

D(ext) flux Figure 13. Results plot for several simulations of D(ext) and F(ext) fluxes Do it by yourself and try to understand why when D(ext) flux is equal or greater than 15 units, no F(Ext) metabolite can be produced in the cell.

Introduction

After having had the first contact with metabolic models, we will go ahead with a genome-scale metabolic model of the cyanobacterium Synechocystis sp. PCC 6803 [Montagud et al., BMC Systems Biology 2010, 4:156], that includes more than 800 metabolic reactions.

After creating the model from the template or accessing as a guest you can click the "Export" button to download the iSyn811 model as a txt file (you can use this file to upload the model, too). Take a look to its content, specially paying special attention to the metabolic pathways structures. Further it is worth to remark that there are three types of reactions in this model:

• uptake reactions, which are normally constrained but can be also unconstrained,
• normal metabolic reactions and
• the biomass equation, which is a set of macromolecules that build the cell structure:

As it can be seen in the figure, the biomass equation is composed by a selection of building blocks to give one unit of biomass (BM) that is transformed into external biomass, also called cell growth (BM out), by the next reaction in the model. The external metabolites that need to be uptaken for the cell metabolism are easy to find as their corresponding uptake reactions appear just after the biomass equation and all external metabolites are displayed with blue colour in WeDesign.

The units of this model are the following:

Simulation of several growing conditions

For the first exercise we will mainly study the cellular growth at three different growth modes, whose corresponding environmental conditions are given in the table:

Table: Initial conditions for growth modes in Synechocystis sp. PCC 6803 glucose tolerant strain

Try to set the constraints from the table and check the resultant growth (reaction _Growth set as Main objective). Note that the constraints have been adjusted to produce similar growth values in the three different scenarios. The carbon intake is equivalent in terms of C atoms. Also, the amount of photons entering the cell were obtained as the minimum necessary to reduce the carbon intake in the autotrophic scenario [Montagud et al., BMC Systems Biology 2010, 4:156].

Maximisation of autotrophic hydrogen production with different nitrogen sources

In this case hydrogen will be generated in autotrophic mode, fixing the growth value as the one obtained in the previous exercise simulation for autotrophic mode and setting hydrogen as Main Objective. Four different simulations with four different nitrogen sources will be evaluated:

• Nitrate (standard conditions)
• Nitrite
• Ammonia
• Urea

All constraints are given in the table:

After having done all simulations a similar graph could be obtained:

Figure 16. Summary results of photohydrogen production in Synechocystis model with different nitrogen sources Why urea and ammonia allow the cell to produce more hydrogen?

Compatible solutes case study

In the last exercise, the effect of three different compatible solute cases on cell growth will be assessed. Compatible solutes are a functional group of low molecular mass, organic compounds that do not disturb cellular metabolism at the high molar concentrations necessary to equilibrate osmotic conditions. So they help protecting the cell from salt stress.

For solving it, you should add the necessary reactions for each solute case, that is synthesis and transport, and then simulate _Growth as Main Objective and _obj as Design Objective. Note that the reaction _obj has the same structure in all cases, therefore you don’t need to add it each time and you can simply modify it, as we saw in the case of the Toy Model. Remember that you should disable the reactions previously added (or constraining them to [0, 0]) to reset the model to the ‘wild type’. The compatible solutes to be studied are Glycine betaine, Trehalose and Octaine. In the case of Trehalose two different pathways can be tried. Below you will find the reactions that are necessary for each simulation.